Arasu Paper PDF

Regional Tribology Conference
Bayview Hotel, Langkawi Island, Malaysia, 22-24 November 2011
Paper Reference ID: RTC 011
SURFACE TEMPERATURE DISTRIBUTION IN A COMPOSITE BRAKE ROTOR
A.A. Adebisi1, M. A. Maleque1 and Q. H Shah2

1
Department of Manufacturing and Materials Engineering,
2
Department of Mechanical Engineering
International Islamic University Malaysia
53100 IIUM Gombak, Kuala Lumpur, Malaysia
e-mail: debisi1@yahoo.com, maleque@iium.edu.my, hqasim@iium.edu.my
ABSTRACT brake rotor play an important role by influencing the

thermal conductivity and heat dissipation during
The prediction of surface temperature for brake rotor is braking. Recent studies have shown that advanced
regarded as an important step in studying the brake composite such as aluminium matrix reinforced with
system performance. The frictional heat generated on silicon carbide particle is a potential material for brake
the rotor surface can influence excessive temperature rotor development due to its thermo-physical
rise which in turn leads to undesirable effects such as properties (Qi, et al., 2001). In a study by Gao and Lin
thermal elastic instability (TEI), premature wear, brake (2002), they observed that considerable evidence has
fluid vaporization (BFV) and thermally excited shown that the contact temperature distribution is an
vibrations (TEV). The purpose of this study is to integral factor influencing the combined effect of load,
investigate the temperature distribution profile for speed, friction coefficient and the thermo-physical and
brake caliper pressure application of 0.5, 1.0 MPa with durability properties of the materials. In another study
a speed of 60km/h braking condition on the disc rotor Lee and Yeo (2000) stated that the uneven distribution
surface. The brake rotor assembly is built by using a 3 of temperature at the surfaces of the rotor could bring
dimensional finite element model of a real car brake about thermal distortion which causes thermal judder
rotor. To verify the simulation results, an experimental and excited vibration.
investigation is carried out. It is believed from the
study that composite brake rotor influences the Finite element (FE) method for brake rotor analysis has
temperature distribution and heat dissipation rate become a preferred method in studying the thermal
which could prevent excessive temperature rise and distribution performance because of its flexibility and
subsequently prolong the service life of the rotor. The diversity in providing solutions to problems involving
finite element method is cost effective and also assists advanced material properties. Chandrupatla and
the automotive industry in producing optimised and Belegundu (2002) stated that temperature distribution
effective brake rotor for thermal distribution analysis. analysis is mostly performed using FE method due to
its powerful tool for numerical solutions for a wide
Keyword: brake rotor, temperature distribution, finite range of engineering problems. Day (1988) conducted
element model, frictional heat a study using FE to predict temperature, wear, pressure
distribution and thermal distortion of a brake drum
1.0 INTRODUCTION which is generated during high pressure brake
application from two different road speed and friction
Brake system is an essential component in the materials. Valvano and Lee (2000) proposed a thermal
automotive industry due to its safety concern to reduce analysis on disc brake based on a combination of
or stop a vehicle on high speed. The braking computer based thermal model and FE based
performance is significantly affected by the techniques to provide reliable method to calculate the
temperature rise in the process of halting the vehicle. temperature rise and distortion under a given brake
Each moment (time step) during the continuous schedule.
braking process gives a different value of temperature
distribution as a result of the frictional heat generated In this paper, the FE model of a real brake rotor
on the rotor surface which can cause high temperature assembly is developed and simulated using the
rise (Qi and Day, 2007; Hwang and Wu, 2010). When commercially available FE software packages, ANSYS
the temperature rise exceeds the critical value for a and LS-prepost respectively. The model is simulated
given material, it leads to undesirable effects in the using a 3D thermo-mechanical coupling model in order
operation of the rotor such as thermal elastic instability to observe the surface temperature distributions profile
(TEI), premature wear, brake fluid vaporization (BFV) for different applied braking condition.
and thermally excited vibrations (TEV) (Gao and Lin,
2002; Kao, et al., 2000). The material properties of the
145
2.0 FINITE ELEMENT MODEL as shown in Figure 1. The FE model is developed

based on the actual Proton Wira 1.3 solid brake rotor
The finite element model of a real brake rotor consists assembly.
of the composite rotor disc and two friction materials
(a) Finite element model (b) Proton Wira brake assembly
Figure 1: Brake rotor assembly
The brake rotor model assembly utilizes up to 9053 degrees of freedom at each node: translations,
solid elements with the rotor element comprising of velocities, and accelerations in the nodal x, y, and z
8787 and the pads 133 elements. The SOLID 164 directions. It gives a reduced one point integration
element type is used for the three-dimensional which saves computer time and robustness in cases of
modeling of the brake rotor solid structures. The large deformations. The description of the brake rotor
element is defined by eight nodes having the following model is given in Table 1.
Table 1: Description of the brake rotor assembly components

Components Number of
Type of Elements Number of nodes
Elements
Rotor Solid 164 8787 7638
Pads Solid 164 133 640
The FE model structure is imported into the LS-prepost the thermal contact conductance as a function of
software in preparation for the implicit dynamic temperature, pressure parameters and contact stiffness.
solution. The contact type is defined as automatic This is to ensure that the temperature distributions on
surface to surface thermal friction for the model which the rotor/pad interface is more significant compared to
defined the mechanical static and dynamic friction other contact interfaces. The rotor is chosen as the
coefficient as a function of temperature. It also defined
146
master surface due to its stiffness, while the friction

materials were chosen as the slave surface. 3.0 THERMAL DISTRIBUTION ANALYSIS
2.1 Boundary Conditions The dissipated energy converted into heat is specified
as all the mechanical energy is converted into thermal
For structural and thermal analysis of the brake rotor energy. Energy dissipated as heat between the surfaces
model, boundary condition is specified; and the distributions are equal between the two
interacting surfaces. Heat is generated on the surfaces
2.1.1 Structural boundary condition between the rotor and pad when the rotor rotates. This
It is specified by imposing nodal motion on the set of could be expressed as (Al-Bahkali and Barber, 2006).
nodes and the motion is prescribed with respect to the
local coordinate system of the brake rotor. The degree q  Vp .............................. (1)
of freedom (DOF) for the boundary prescribed motion
specifies that x/y DOF for node rotating about the z
axis is at a location specified in the x-y plane. The SPC where  is the friction coefficient, V is the sliding
set specifies the constraints at the nodal single points. velocity of the rotor and p is the contact pressure at
the interface, q is the amount of heat generated by
2.1.2 Thermal boundary condition
friction. For other regions on the rotor and pad exposed
Boundary temperature condition for the set of nodes is
to the environment, it is assumed that the heat
specified for coupled thermal/ structural analysis of the
exchange is transferred through convection process.
brake rotor by the load curve ID for temperature versus
Therefore, convection surface boundary condition is
time interval.
applied. This can be expressed as:
T
k  h T   T  0, t  
x .............. (2)
where h is convection heat transfer coefficient, T  is

atmosphere temperature and T  0, t  is the current
temperature of the node.
4.0 SIMULATION AND EXPERIMENTATION

Figure 2: Thermal boundary condition
In the present study, a proton wira with vehicle curb
2.2 Load Application weight of 1250kg is utilised, the friction and drag
coefficient of the contact pair is 0.35 and 0.30
Load is applied to the rotor model structure in respectively with an initial temperature of 35C. The
preparation for explicit dynamic solution. The pressure rotor material for the study is 20 wt% MPS-SiC AMC.
is defined using load segment keyword which is The dimension and material properties of the brake
applied to the faces of the model, on top of the rotor and pads are listed in Table 2.
appropriate solid elements (as rigid body). The faces
are defined with segments and the load is defined with Table 2
the load curve number. The load curve is specified Material property and dimension of brake rotor and pad
with a well defined load direction before it is then Rotor Pad
applied on the brake rotor model as Inner radius (mm) 135 155
shown in the Figure 3. Outer radius (mm) 230 221
Thickness (mm) 15 10
Density (kg/m3) 2.903 2.595
Specific heat (Nm/kgK) 845 1465
Thermal conductivity (Nm/s°Cm) 170 1.212
Young`s modulus (GPa) 113 22
Poisson`s ratio 0.24 0.25
Tensile strength (MPa) 178 -
The vehicle speed is 60km/h during the static running

test carried out in the automotive laboratory for varied
brake pressure application. Brake pressure of 0.5, 1.0
Figure 3: Load application on brake rotor and pad MPa is applied on the pad through the caliper piston to
generate the pressure which is monitored with a
147
pressure gauge from the caliper valve (nipple). The rotor where it becomes constant. The rotational speed
model is symmetrical about the work surface of the of the rotor during contact with the pad develops
friction contact pair which is defined to carry out frictional heat until the temperature gradually
simulation for the temperature distribution profile. increases. After which the rotation of the rotor
Based on the 3D thermo-mechanical coupling becomes constant and the thermal analysis continue
technique, the analysis generated for the braking until the end of the simulation. The mid distance region
process was presented for temperature versus time of contact between the rotor and pad is analysed for
interval. To verify the simulation results, an temperature distribution profile on the rotor surface.
experimental investigation was carried out for the
AMC brake rotor temperature distribution and also Figure 4 shows a temperature profile for pressure
compared with the conventional cast iron brake disc application of 0.5MPa and figure 5 gives the
rotor. corresponding mid radial temperature distribution plot,
the temperature gradually increases to 78°C for a time
5.0 RESULTS AND DISCUSSION period of 20 ms. Figure 6 shows the temperature
profile for pressure of 1MPa with a temperature rise of
Several assumptions were taken into consideration 147.7°C for 20 ms and figure 7 gives the mid radial
when performing the thermal analysis. The applied temperature distribution plot respectively. From the
brake pressure is assumed to be uniformly distributed surface temperature profile plots, it shows that the
on the brake pads during operation. The coefficient of temperature increases and decreases at certain region at
friction is assumed to remain constant throughout the the same time interval. The increase in temperature
analysis. The material and thermal properties are results from the rotor contact with the pad, and when
homogeneous and invariant with the temperature. The the rotor slides away from the pad the temperature will
wear affect is also neglected. slightly drop. The reason for this is as a result of the
cooling effect through heat transfer process
Brake pressure is applied directly on the pads through (conduction) which also depends on the material
the caliper piston until it makes contact with the brake properties of the rotor.
Figure 4: Surface temperature distribution profile for 0.5MPa
Figure 5: Mid radial temperature distribution plot for 0.5MPa
148
From the result, the time history curve for the rotor
surface shows that temperature increases linearly
before dropping which indicates that increase in brake
pressure increases the surface temperature of the rotor. Cast Iron Rotor
The study also found that higher temperature occurs at
the center of the rotor surface and it spreads to the
circumferential direction. The inner portion of the rotor
remains the warmest section. The AMC brake rotor
properties exhibited better distribution of temperature Figure 8: Thermograph analysis for average surface
which reduced the localization of heat generation temperature profile for AMC and cast iron brake rotor.
thereby influencing the hot spot and thermal elastic
instability (Khai, et al., 2007). Experimental test was At a pressure of 0.5MPa the average surface
conducted for both AMC and cast iron brake rotors. temperature for both rotors is slightly different, but as
Figure 8 shows the profile analysis for both rotors. the pressure applied is increased to 1 MPa the
temperature gap increases, this is due to the material
P = 0.5MPa P = 1MPa properties application. The surface temperature profile
T = 20 secs T = 20 secs of the rotor measured in the experimentation test for
0.5 and 1MPa is shown in Figure 8. This shows that
the simulation results for the AMC brake rotor are in
good agreement with the experimental values.
AMC Rotor
CONCLUSIONS
149
The present study investigated the surface temperature Chandrupatla, T.R. and Belegundu, A. D. (2002)
distribution analysis of the AMC brake rotor. The LS- Introduction to finite elements in engineering.
prepost (LS-Dyna) finite element software package is 3rd edition Prentice Hall.
utilized to predict the temperature distribution on the Day, A. (1988). An analysis of speed, temperature and
rotor surface. Long duration investigation is limited performance characteristics of automotive
due to hardware limitation and time constraint, drum brakes. Journal of tribology, 110: 298.
although results generated from the investigation is Gao, C. and Lin, X. (2002). Transient temperature field
adequate to observe some relevant characteristics of analysis of a brake in a non-axisymmetric
temperature distribution profile. Moreover, the three-dimensional model. Journal of materials
following conclusions can be made from the study; processing technology, 129 (1-3): 513-517.
1. Successful development of AMC brake rotor Hwang, P. and Wu, X. (2010) Investigation of
through experimental and FE model analysis. temperature and thermal stress in ventilated
2. The AMC brake rotor exhibited better disc brake based on 3D thermo-mechanical
distribution of temperature which reduces the coupling model. Journal of mechanical
localization of heat generation thereby science and technology, 24(1): 81-84.
influences thermoelastic instability TEI, Kao, T., Richmond, J. and Douarre, A. (2000) Brake
premature wear and thermally excited disc hot spotting and thermal judder: an
vibrations TEV. experimental and finite element study.
3. Properties of the AMC rotor shows improved International Journal of Vehicle Design,
cooling effect due to its high thermal 23(3): 276-296.
conductivity when compared to conventional Khai, L. C., and Abu Bakar, A. and Abdullah, M.
cast iron properties. S. (2007). Prediction of temperature
4. Both the experimental and simulated results distributions in a disc assembly using the
for the AMC rotor are in good agreement. finite element method. 5th Malaysian Friction
ACKNOWLEDGEMENT Materials.
The authors acknowledge the support of the Qi, H.S. and Day, A. J. (2007). Investigation of
department of Manufacturing and Materials disc/pad interface temperatures in friction
Engineering, International Islamic University Malaysia braking. Wear, 262 (5-6): 505-513.
and also grateful to the Research Management Centre, Qi, H., Fan, Y. and Ding, Z. (2001). Application of the
International Islamic University Malaysia (IIUM), for particular reinforced aluminum composite to
financial support to conduct this research work under automobile brake rotors. Journal of the Hebei
project EDW B 0906-332. Academy of Sciences, 2.
Sangkook, L. and Taein, Y. (2000). Temperature and
REFERENCES coning analysis of brake rotor using an
Al-Bahkali, E. A. and Barber, J. R. (2006). Nonlinear axisymmetric finite element technique.
Steady State Solution for a Thermoelastic Science and Technology, 17-22.
Sliding System Using Finite Element Method. Valvano, T., Lee, K. and Systems, D. A. (2000). An
Journal of Thermal Stresses, 29 (2): 153-168 analytical method to predict thermal distortion
of a brake rotor. SAE transactions, 109(6):
566-571.
150
SURFACE TEMPERATURE DISTRIBUTION IN A COMPOSITE BRAKE ROTOR
A.A. Adebisi1, M. A. Maleque1 and Q. H Shah2

1
Department of Manufacturing and Materials Engineering,
2
Department of Mechanical Engineering
e-mail: debisi1@yahoo.com, maleque@iium.edu.my, hqasim@iium.edu.my
ABSTRACT brake rotor play an important role by influencing the

thermal conductivity and heat dissipation during
The prediction of surface temperature for brake rotor is braking. Recent studies have shown that advanced
regarded as an important step in studying the brake composite such as aluminium matrix reinforced with
system performance. The frictional heat generated on silicon carbide particle is a potential material for brake
the rotor surface can influence excessive temperature rotor development due to its thermo-physical
rise which in turn leads to undesirable effects such as properties (Qi, et al., 2001). In a study by Gao and Lin
thermal elastic instability (TEI), premature wear, brake (2002), they observed that considerable evidence has
fluid vaporization (BFV) and thermally excited shown that the contact temperature distribution is an
vibrations (TEV). The purpose of this study is to integral factor influencing the combined effect of load,
investigate the temperature distribution profile for speed, friction coefficient and the thermo-physical and
brake caliper pressure application of 0.5, 1.0 MPa with durability properties of the materials. In another study
a speed of 60km/h braking condition on the disc rotor Lee and Yeo (2000) stated that the uneven distribution
surface. The brake rotor assembly is built by using a 3 of temperature at the surfaces of the rotor could bring
dimensional finite element model of a real car brake about thermal distortion which causes thermal judder
rotor. To verify the simulation results, an experimental and excited vibration.
investigation is carried out. It is believed from the
study that composite brake rotor influences the Finite element (FE) method for brake rotor analysis has
temperature distribution and heat dissipation rate become a preferred method in studying the thermal
which could prevent excessive temperature rise and distribution performance because of its flexibility and
subsequently prolong the service life of the rotor. The diversity in providing solutions to problems involving
finite element method is cost effective and also assists advanced material properties. Chandrupatla and
the automotive industry in producing optimised and Belegundu (2002) stated that temperature distribution
effective brake rotor for thermal distribution analysis. analysis is mostly performed using FE method due to
its powerful tool for numerical solutions for a wide
Keyword: brake rotor, temperature distribution, finite range of engineering problems. Day (1988) conducted
element model, frictional heat a study using FE to predict temperature, wear, pressure
distribution and thermal distortion of a brake drum
1.0 INTRODUCTION which is generated during high pressure brake
application from two different road speed and friction
Brake system is an essential component in the materials. Valvano and Lee (2000) proposed a thermal
automotive industry due to its safety concern to reduce analysis on disc brake based on a combination of
or stop a vehicle on high speed. The braking computer based thermal model and FE based
performance is significantly affected by the techniques to provide reliable method to calculate the
temperature rise in the process of halting the vehicle. temperature rise and distortion under a given brake
Each moment (time step) during the continuous schedule.
braking process gives a different value of temperature
distribution as a result of the frictional heat generated In this paper, the FE model of a real brake rotor
on the rotor surface which can cause high temperature assembly is developed and simulated using the
rise (Qi and Day, 2007; Hwang and Wu, 2010). When commercially available FE software packages, ANSYS
the temperature rise exceeds the critical value for a and LS-prepost respectively. The model is simulated
given material, it leads to undesirable effects in the using a 3D thermo-mechanical coupling model in order
operation of the rotor such as thermal elastic instability to observe the surface temperature distributions profile
(TEI), premature wear, brake fluid vaporization (BFV) for different applied braking condition.
and thermally excited vibrations (TEV) (Gao and Lin,
2002; Kao, et al., 2000). The material properties of the
145
2.0 FINITE ELEMENT MODEL as shown in Figure 1. The FE model is developed

based on the actual Proton Wira 1.3 solid brake rotor
The finite element model of a real brake rotor consists assembly.
of the composite rotor disc and two friction materials
(a) Finite element model (b) Proton Wira brake assembly
Figure 1: Brake rotor assembly
The brake rotor model assembly utilizes up to 9053 degrees of freedom at each node: translations,
solid elements with the rotor element comprising of velocities, and accelerations in the nodal x, y, and z
8787 and the pads 133 elements. The SOLID 164 directions. It gives a reduced one point integration
element type is used for the three-dimensional which saves computer time and robustness in cases of
modeling of the brake rotor solid structures. The large deformations. The description of the brake rotor
element is defined by eight nodes having the following model is given in Table 1.
Table 1: Description of the brake rotor assembly components

Components Number of
Type of Elements Number of nodes
Elements
Rotor Solid 164 8787 7638
Pads Solid 164 133 640
The FE model structure is imported into the LS-prepost the thermal contact conductance as a function of
software in preparation for the implicit dynamic temperature, pressure parameters and contact stiffness.
solution. The contact type is defined as automatic This is to ensure that the temperature distributions on
surface to surface thermal friction for the model which the rotor/pad interface is more significant compared to
defined the mechanical static and dynamic friction other contact interfaces. The rotor is chosen as the
coefficient as a function of temperature. It also defined
146
master surface due to its stiffness, while the friction

materials were chosen as the slave surface. 3.0 THERMAL DISTRIBUTION ANALYSIS
2.1 Boundary Conditions The dissipated energy converted into heat is specified
as all the mechanical energy is converted into thermal
For structural and thermal analysis of the brake rotor energy. Energy dissipated as heat between the surfaces
model, boundary condition is specified; and the distributions are equal between the two
interacting surfaces. Heat is generated on the surfaces
2.1.1 Structural boundary condition between the rotor and pad when the rotor rotates. This
It is specified by imposing nodal motion on the set of could be expressed as (Al-Bahkali and Barber, 2006).
nodes and the motion is prescribed with respect to the
local coordinate system of the brake rotor. The degree q  Vp .............................. (1)
of freedom (DOF) for the boundary prescribed motion
specifies that x/y DOF for node rotating about the z
axis is at a location specified in the x-y plane. The SPC where  is the friction coefficient, V is the sliding
set specifies the constraints at the nodal single points. velocity of the rotor and p is the contact pressure at
the interface, q is the amount of heat generated by
2.1.2 Thermal boundary condition
friction. For other regions on the rotor and pad exposed
Boundary temperature condition for the set of nodes is
to the environment, it is assumed that the heat
specified for coupled thermal/ structural analysis of the
exchange is transferred through convection process.
brake rotor by the load curve ID for temperature versus
Therefore, convection surface boundary condition is
time interval.
applied. This can be expressed as:
T
k  h T   T  0, t  
x .............. (2)
where h is convection heat transfer coefficient, T  is

atmosphere temperature and T  0, t  is the current
temperature of the node.
4.0 SIMULATION AND EXPERIMENTATION

Figure 2: Thermal boundary condition
In the present study, a proton wira with vehicle curb
2.2 Load Application weight of 1250kg is utilised, the friction and drag
coefficient of the contact pair is 0.35 and 0.30
Load is applied to the rotor model structure in respectively with an initial temperature of 35C. The
preparation for explicit dynamic solution. The pressure rotor material for the study is 20 wt% MPS-SiC AMC.
is defined using load segment keyword which is The dimension and material properties of the brake
applied to the faces of the model, on top of the rotor and pads are listed in Table 2.
appropriate solid elements (as rigid body). The faces
are defined with segments and the load is defined with Table 2
the load curve number. The load curve is specified Material property and dimension of brake rotor and pad
with a well defined load direction before it is then Rotor Pad
applied on the brake rotor model as Inner radius (mm) 135 155
shown in the Figure 3. Outer radius (mm) 230 221
Thickness (mm) 15 10
Density (kg/m3) 2.903 2.595
Specific heat (Nm/kgK) 845 1465
Thermal conductivity (Nm/s°Cm) 170 1.212
Young`s modulus (GPa) 113 22
Poisson`s ratio 0.24 0.25
Tensile strength (MPa) 178 -
The vehicle speed is 60km/h during the static running

test carried out in the automotive laboratory for varied
brake pressure application. Brake pressure of 0.5, 1.0
Figure 3: Load application on brake rotor and pad MPa is applied on the pad through the caliper piston to
generate the pressure which is monitored with a
147
pressure gauge from the caliper valve (nipple). The rotor where it becomes constant. The rotational speed
model is symmetrical about the work surface of the of the rotor during contact with the pad develops
friction contact pair which is defined to carry out frictional heat until the temperature gradually
simulation for the temperature distribution profile. increases. After which the rotation of the rotor
Based on the 3D thermo-mechanical coupling becomes constant and the thermal analysis continue
technique, the analysis generated for the braking until the end of the simulation. The mid distance region
process was presented for temperature versus time of contact between the rotor and pad is analysed for
interval. To verify the simulation results, an temperature distribution profile on the rotor surface.
experimental investigation was carried out for the
AMC brake rotor temperature distribution and also Figure 4 shows a temperature profile for pressure
compared with the conventional cast iron brake disc application of 0.5MPa and figure 5 gives the
rotor. corresponding mid radial temperature distribution plot,
the temperature gradually increases to 78°C for a time
5.0 RESULTS AND DISCUSSION period of 20 ms. Figure 6 shows the temperature
profile for pressure of 1MPa with a temperature rise of
Several assumptions were taken into consideration 147.7°C for 20 ms and figure 7 gives the mid radial
when performing the thermal analysis. The applied temperature distribution plot respectively. From the
brake pressure is assumed to be uniformly distributed surface temperature profile plots, it shows that the
on the brake pads during operation. The coefficient of temperature increases and decreases at certain region at
friction is assumed to remain constant throughout the the same time interval. The increase in temperature
analysis. The material and thermal properties are results from the rotor contact with the pad, and when
homogeneous and invariant with the temperature. The the rotor slides away from the pad the temperature will
wear affect is also neglected. slightly drop. The reason for this is as a result of the
cooling effect through heat transfer process
Brake pressure is applied directly on the pads through (conduction) which also depends on the material
the caliper piston until it makes contact with the brake properties of the rotor.
148
From the result, the time history curve for the rotor
surface shows that temperature increases linearly
before dropping which indicates that increase in brake
pressure increases the surface temperature of the rotor. Cast Iron Rotor
The study also found that higher temperature occurs at
the center of the rotor surface and it spreads to the
circumferential direction. The inner portion of the rotor
remains the warmest section. The AMC brake rotor
properties exhibited better distribution of temperature Figure 8: Thermograph analysis for average surface
which reduced the localization of heat generation temperature profile for AMC and cast iron brake rotor.
thereby influencing the hot spot and thermal elastic
instability (Khai, et al., 2007). Experimental test was At a pressure of 0.5MPa the average surface
conducted for both AMC and cast iron brake rotors. temperature for both rotors is slightly different, but as
Figure 8 shows the profile analysis for both rotors. the pressure applied is increased to 1 MPa the
temperature gap increases, this is due to the material
P = 0.5MPa P = 1MPa properties application. The surface temperature profile
T = 20 secs T = 20 secs of the rotor measured in the experimentation test for
0.5 and 1MPa is shown in Figure 8. This shows that
the simulation results for the AMC brake rotor are in
good agreement with the experimental values.
AMC Rotor
CONCLUSIONS
149
The present study investigated the surface temperature Chandrupatla, T.R. and Belegundu, A. D. (2002)
distribution analysis of the AMC brake rotor. The LS- Introduction to finite elements in engineering.
prepost (LS-Dyna) finite element software package is 3rd edition Prentice Hall.
utilized to predict the temperature distribution on the Day, A. (1988). An analysis of speed, temperature and
rotor surface. Long duration investigation is limited performance characteristics of automotive
due to hardware limitation and time constraint, drum brakes. Journal of tribology, 110: 298.
although results generated from the investigation is Gao, C. and Lin, X. (2002). Transient temperature field
adequate to observe some relevant characteristics of analysis of a brake in a non-axisymmetric
temperature distribution profile. Moreover, the three-dimensional model. Journal of materials
following conclusions can be made from the study; processing technology, 129 (1-3): 513-517.
1. Successful development of AMC brake rotor Hwang, P. and Wu, X. (2010) Investigation of
through experimental and FE model analysis. temperature and thermal stress in ventilated
2. The AMC brake rotor exhibited better disc brake based on 3D thermo-mechanical
distribution of temperature which reduces the coupling model. Journal of mechanical
localization of heat generation thereby science and technology, 24(1): 81-84.
influences thermoelastic instability TEI, Kao, T., Richmond, J. and Douarre, A. (2000) Brake
premature wear and thermally excited disc hot spotting and thermal judder: an
vibrations TEV. experimental and finite element study.
3. Properties of the AMC rotor shows improved International Journal of Vehicle Design,
cooling effect due to its high thermal 23(3): 276-296.
conductivity when compared to conventional Khai, L. C., and Abu Bakar, A. and Abdullah, M.
cast iron properties. S. (2007). Prediction of temperature
4. Both the experimental and simulated results distributions in a disc assembly using the
for the AMC rotor are in good agreement. finite element method. 5th Malaysian Friction
ACKNOWLEDGEMENT Materials.
The authors acknowledge the support of the Qi, H.S. and Day, A. J. (2007). Investigation of
department of Manufacturing and Materials disc/pad interface temperatures in friction
Engineering, International Islamic University Malaysia braking. Wear, 262 (5-6): 505-513.
and also grateful to the Research Management Centre, Qi, H., Fan, Y. and Ding, Z. (2001). Application of the
International Islamic University Malaysia (IIUM), for particular reinforced aluminum composite to
financial support to conduct this research work under automobile brake rotors. Journal of the Hebei
project EDW B 0906-332. Academy of Sciences, 2.
Sangkook, L. and Taein, Y. (2000). Temperature and
REFERENCES coning analysis of brake rotor using an
Al-Bahkali, E. A. and Barber, J. R. (2006). Nonlinear axisymmetric finite element technique.
Steady State Solution for a Thermoelastic Science and Technology, 17-22.
Sliding System Using Finite Element Method. Valvano, T., Lee, K. and Systems, D. A. (2000). An
Journal of Thermal Stresses, 29 (2): 153-168 analytical method to predict thermal distortion
of a brake rotor. SAE transactions, 109(6):
566-571.
150
WEAR PROPERTIES OF Fe-C-Al CAST IRON AND CONVENTIONAL CAST

IRON –A COMPARATIVE STUDY
M. A. Maleque*
Department of Manufacturing and Materials Engineering

Faculty of Engineering, International Islamic University Malaysia
PO Box 10, 50728 Kuala Lumpur, Malaysia
*Corresponding author / email: maleque@iium.edu.my
ABSTRACT field of machine parts and components such as

roller, roller shell, lathe bench, agricultural tools
The conventional gray cast iron (also known as etc. In the recent years, the demand for gray cast
Fe-C-Si cast iron) exhibitss good properties, iron in different applications are increasing and
such as strength, hardness, corrosion and therefore, there is a need to develop new grades
oxidation resistance but it’s properties of gray cast iron which can provide better
especially wear and friction detoriorate physical, mechanical and wear properties.
significantly at higher temperature. In order to
improve the wear properties of this Gray cast iron is one of the most easily
conventional cast iron, an attemped has been cast of all metals in the foundry. It has the
made to replace Si by Al. In the present study, lowest pouring temperature of the ferrous
the wear test was conducted on both metals, which is reflected in its high fluidity and
conventional cast iron (CI) and Fe-C-Al cast its ability to be cast into intricate shapes. As a
iron using universal pin-on-disk (POD) result of a custom during final stages of
machine. The wear track profile and damage solidification, it has very low solid shrinkage so
surface were characterized using optical that sound castings are readily obtainable. For
profilometer and scanning electron microscopy the majority of applications, gray iron is used in
(SEM) respectively. The results showed that the its as-cast condition, thus simplifying
lower wear rate was found for Fe-C-Al CI production. Gray iron has excellent machining
compared to conventional Fe-C-Si CI and qualities producing easily disposed of chips and
abrasive type wear morphology was observed in yielding a better surface finish. The resistance
both types of cast iron materials. The results of gray iron to scoring and galling with proper
also showed that the friction coeficient value of matrix and graphite structure is universally
Fe-C-Al CI was lower than conventional cast recognized (Krause, 1969). However, the
iron. It can be concluded that the overall wear mechanical and chemical properties of cast iron
and friction properties of Fe-C-Al CI was better can be widely changed by alloy addition, heat
than conventional cast iron. treatment or controlling the solidification of the
alloy (Bartocha et. al., 2005; Xing et al., 2007).
Keywords: Sliding; Two-body abrasion; Cast The wear property of gray cast iron mainly
Iron; Surface Analysis; Profilometry. depends on the properties of the matrix and
hence, by modifying the matrices of the gray
1. INTRODUCTION cast iron it is possible to change the wear
behaviour of the cast iron (Terheci, et al., 1995).
Gray cast iron is one of the oldest cast ferrous Riahi and Alpas (2003) and Prasad (2007) have
products. In spite of competition from newer suggested that addition of aluminium in place of
materials and their energetic promotion, gray silicon might form fine ferrite or ferrite pearlite
iron is still used for those applications where its which in tern increase the toughness and
properties have proved it to be the most suitable strength of the cast iron. Low temperature
material available. Next to wrought steel, gray damping behaviour of cast iron with aluminium
iron is the most widely used metallic material addition has been studied by Liu et al., (2005).
for engineering purposes including automotive They found that the addition of aluminium in
engine component material due to its multi- gray iron can modify the matrix with flake
variety properties such as, good hardness and which contribute to the improvement in the
strength, higher damping capacity and better structure and properties compared to
tribological properties at ambient temperature. conventional gray iron. Not much information is
The other applications of this material are in the available in literature on the comparative study
151
of wear properties bewteen conventional gray ASTM G 99-95a standard. Before placing the
CI and Fe-C-Al CI. Therefore, the main aim of sample into the specimen chamber, it was
this work is to study the wear behaviour of Fe- cleaned with acetone to remove oil, grease or
C-Al gray cast iron and conventional gray iron dust from the surface for each test. The wear
using universal pin-onodisk (POD) machine. and friction data are sent directly to a personal
computer via data acquisition system. After test,
worn surfaces were cleaned lightly using
2. EXPERIMENTAL DETAILS acetone solution for surface characterization.
2.1 Casting Process
The moulding materials for casting are:

synthetic silica sand, cold set asphalt resin and
alpha cure hardener. In order to prepare a
mould, first silica sand was mixed with 2% resin
for 5 minutes using a blender then added 1%
cold set hardener and mix it thoroughly for
another 1 minute. Finally, this freshly sand
mixture was used to make a mould on wooden
pattern.
An induction furnace was used for melting the

raw materials such as pig iron, mild steel,
limestone and ferrosilicon. Firstly, pig iron and Figure 1 A schematic diagram of pin on disc
mild steel were charged together in the furnace wear tester under dry sliding condition.
and heated up to about 1350 0C. Then,
commercially available flux material
(limestone) was added into the melt and stirred 2.3 Surface Characterization
thoroughly. The slag was removed from the top
of the melt and ferrosilicon alloy was added into Surface analysis on wear track after wear
the melt manually in order to adjust the test of Fe-C-Al CI and conventional CI were
composition of the cast iron. The above step performed using “Veeco Wyko NT1100”
was repeated by adding the flux material to optical profilometry and presented in both 2D
ensure the removal of slag from melt by keeping and 3D profiles of the wear track. The wear
the melt in the furnace in order to have the high worn surface was studied using SEM (Philips
quality of cast product. Finally, the liquid metal model XL20) in order to understand the wear
was poured into the cold set resin bonded sand morphology and identify the nature and the
mould at about 1300 0C and solidified it to room mechanism of wear after wear and friction test.
temperature. After casting, the samples were
prepared for wear study.
3. RESULTS AND DISCUSSION
2.2 Wear and Friction Test
3.1 Wear Properties of Fe-C-Al and
Conventional Cast Irons
The dry sliding wear test was performed using a
universal wear testing machine (CSM High The wear rate of both Fe-C-Al and conventional
Temperature Tribometer) as shown in Fig. 1. (Fe-C-Si) cast irons is shown in Fig 2. From
Both Fe-C-Al and conventional CI samples figure, it is noticed that the conventional cast
were used as a rotating disk, whereas a 6 mm iron showed higher wear rate compared to Fe-C-
diameter alumina ball pin (with the hardness Al cast iron. On the contrary, the increase rate
value of ~1500, Vickers kgf.mm-2) was used as of wear for Fe-C-Al cast iron was minimum
a counter-part material. The hardness of Fe-C- with R2 value of 0.9997 which indicate the
Al CI and conventional CI are 240 BHN and steady state and uniform wear rate with higher
190 BHN respectively. The disk dimension was slope. In fact, the wear rate of conventional
30×30×10 mm and finely polished before test. (Fe-C-Si) cast iron varies from 12.42×10-5
All the tests were carried out at 5 N load and at mm3/m/N to 98.88×10-5 mm3/m/N, whereas the
a fixed sliding spped of 25 cm/s at 56% humid wear rate of Fe-C-Al cast iron varies only from
atmosphere. The test duration for each test was 3.3203×10-5 mm3/m/N to 8.1550×10-5
5 min and the test was performed according to 3
mm /m/N. However, the Si containing cast iron
152
has more graphite flakes and hence matrix will C-Si CI compared to Fe-C-Al CI. This again
get more chance to be separated than less shows the highrer wear protection of Fe-C-Al
graphite flaky cast iron which is Fe-C-Al CI. CI compared to conventional CI and also well
The highly dense graphite flakes in presence of agreed with the previous explanation on wear
higher Si finally concentrated the stress during rate in Fig. 2.
wear test (Hatate et al., 2001) resulting peeling
off the Fe-C-Si cast iron materials as shown in
Fig 5 (a). Therefore, it can be concluded that
the Fe-C-Al cast iron is more wear protective
than conventional Fe-C-Si cast iron.
(a)
Figure 2 Wear rate of Fe-C-Al and conventional

Fe-C-Si cast irons under dry sliding condition.
The speed is 25 cm/s; load 5N; duration 5 min.
(b)
Figure 4 Wear tracks of Fe-C-Si cast iron under
optical profilometer: (a) wear track profile and
(b) corresponding depth of wear track.
3.2 Wear Mechanism of Damage Surface
The wear worn or damage surface of both cast

(a)
iron materials under SEM are shown in Fig 5
represent the morphology of the wear. It shows
that conventional Fe-C-Si cast iron has larger
plastic flow, scar, more debris and cutt-off or
peel-off surface resembling delamination on the
surface indicating severe wear of the Fe-C-Si
cast iron material.
However, worn surface of Fe-C-Al CI

shows less plastic flow and wear scar indicating
(b) less severe wear on the damage surface. When
Figure 3 Wear tracks of Fe-C-Al cast iron at hard asperity on the surface of a material or a
room temperature under optical profilometer: hard particle is entrapped between two surfaces,
(a) wear track profile and (b) corresponding plastically deformed or cut off surface may
depth of wear track. occur due to the sliding motion, resulting
abrasive wear. In this study, both cast iron
The surface area of wear track profiles under materials surface were cut off due to sliding
optical profilometer for Fe-C-Al CI and motion and thus abrasive type of wear occurred
conventional gray cast iron materials are shown in this investigation.
in Figs 3 and 4 respectively. From both wear
track (3D view) and depth of wear track (2D Abrasive wear property also depends on
view) profiles it can be seen that the depth of oxidation characteristics of the materials. When
the wear track increases for the conventional Fe- debris forms during wear test, it is clear that
153
oxidation occurs in that place and every moment

virgin materials are exposed. If the formation of
more debris and scar are higher, the wear rate
will also be higher. The conventional Fe-C-Si
cast iron shows larger scar and debris which
indicates higher wear as can be seen in Fig. 2 as
well.
Figure 6 Friction coefficient of both

conventional Fe-C-Si CI and Fe-C-Al cast iron.
4. CONCLUSIONS
(a) Conventional (Fe-C-Si) CI
The wear properties of Fe-C-Al cast iron and
conventional (Fe-C-Si) cast iron were
investigated and the following conclusions can
be drawn from this study:
(i) The wear rate of Fe-C-Al CI is
lower that Fe-C-Si CI.
(ii) The depth of wear track for Fe-C-
Al CI is lower than Fe-C-Si CI
material.
(iii) Abrasive type of wear morphology
was observed in both types of cast
(b) Fe-C-Al CI iron material, however, Fe-C-Si CI
shows more plastic flow and
Figure 5 Wear damage surface of (a) dilamination compared to Fe-C-Al
Conventional Fe-C-Si CI and (b) Fe-C-Al CI CI.
under SEM. (iv) The friction coeficient of both type
of cast irons increase with
increasing the time showing three
3.3 Friction Coefficient regime of frictional behaviour.
(v) Overall results advocate to use Fe-
Friction coefficients of both conventional C-Al cast iron material where wear
Fe-C-Si and Fe-C-Al cast irons at room and friction reristance properties
temperature are plotted with respect to time in are required.
Fig 6. It shows that there are three stages of
friction. In first stage, the friction coefficient is
increased due to formation of tribochemical Acknowledgement
film; in the second stage is steady state where Author is thankful to the Kulliyyah of
the film remains stable and finally, the friction Engineering, International Islamic University
coefficient increases when the film breaks Malaysia for providing lab facilities. The
down. The same trend was also reported by technical support from Br Husni kamal is also
Keller (2007). The friction coefficient for both acknowledged.
types of cast iron shows the same trend and their
values are slightly different from each other.
This may be due to the same operating speed REFERENCES
and ambient room temperature. In a fixed speed,
the same amount of graphite was smeared on Bartocha, D. Janerka K. and Suchon, J. 2005,
the worn track to keep the coefficient of friction Charge materials and technology of
almost similar. The similar phenomena was also melt and structure of gray cast iron, J.
reported by Ghaderi and his coworkers (2003). of Mat Proc. Tech., 162–163, 465-470.
154
Ghaderi, A. R. Ahmadabadi, M. N. and

Ghasemi, H. M. 2003, Effect of
graphite morphologies on the
tribological behavior of austempered
cast iron, Wear, 255, 410-416.
Hatate, M. Shiota, T. Takahashi, N. and

Shimizu, K. 2001, Influences of
graphite shapes on wear characteristics
of austempered cast iron, Wear,
Volume 251, 885-889.
Keller, J. Fridrici, V. Kapsa, P. Vidaller, S. and

Huard, J. F. 2007, Influence of
chemical composition and
microstructure of grey cast iron on
wear of heavy duty diesel engines
cylinder liners, Wear, 263, 1158-1165
Krause, D. E., 1969, "Gray Iron-A Unique

Engineering Material" Gray, Ductile,
and Malleable Iron Castings-Current
Capabilities, ASTM STP 455,
American Society for Testing and
Materials, Philadelphia, 3-28.
Liu, X. Takamori, S. Osawa, Y. and Yin, F.

2005, Low temperature damping
behaviour of cast iron with aluminium
addition, J. of Material Science, 40,
1773-1775.
Prasad, B. K. 2007, Sliding wear response of

cast iron as influenced by micro
structural features and test condition,
Mat. Sc. and Eng.: A, 456, 373.
Riahi, A. R. and Alpas, A. T. 2003, Wear map

for grey cast iron, Wear, 255, 401.
Terheci, M. Manory, R. R. and Hensler, J. H.

1995, The friction and wear of
automotive gray cast iron under dry
sliding condition part 1: relationship
between wear loss and testing
parameters, Wear, 180, 73-78.
Xing, J. Wang, Q. and Chang, J. 2007,

Microstructure of martensite / bainite
dual-phase gray cast iron and its
strengthening mechanism, ISIJ
International, 47, 1776-1780.
155
TRIBOLOGICAL STUDY OF Al-BASED NATURAL FIBRE REINFORCED

COMPOSITE BRAKE PAD MATERIALS
M. A. Malequea*, A. R. Zamania, R. J. Talibb

a*
Department of Manufacturing and Materials Engineering, International Islamic University Malaysia,
Kuala Lumpur, Malaysia; email: maleque@iium.edu.my
b
Advanced Materials Centre, SIRIM Berhad, 09000 Kulim, Kedah, Malaysia
E-mail: talibria@sirim.my
ABSTRACT (Talib et al., 2003). Brake system is used to

slow down and finally stop a moving motor
Tribological behavior of Al–based natural fiber vehicle and to hold vehicle stationary when in
reinforced composites (NFC) brake pad parking position. Friction is required in the
materials was studied using CHASE braking system even though it might cause wear
Dynamometer machine in accordance with of lining material and counter face component.
Society of Automotive Engineer standard Friction and wear characteristics depend on the
procedure SAE J661. In this study, the NFC brake design, the content of friction materials
sample with a size of 25 mm x 25 mm x 6 mm and its behavior in service. Brake pad materials
was fabricated using powder metallurgy route are mainly composed of more than five different
for friction and wear test. Five different materials and can be categorized into four main
laboratory formulations such as S1, S2, S3, S4 constituents, namely fibre, resin, modifier, and
and S5 were prepared with varying coir fiber filler (Talib, 2001). In the brake pads material
contents from 0, 5, 10, 15, and 20 % volume generally asbestos fibers are embedded in
fractions along with binder, friction modifier, polymeric matrix along with several other
abrasive material and solid lubricant. ingredients. The use of asbestos fiber is
Characterization techniques such as SEM and decreasing due to its carcinogenic nature
CHASE Dynamometer machine are used to (Langer, 2002). No single fibre is known that
characterize and analyze the properties of brake function effectively by itself as an asbestos
pad materials. Out of the five formulations, S2 replacement in friction materials.
and S3 showed better properties in terms of
wear and friction compared to others. Current research towards the eco-friendly brake
Formulation S2 exhibits better thermal stability pad material has shown promising interest for
and wear resistance. The microstructure reveals the use of natural fibres such as, hemp and coir
uniform distribution of constituents materials fibre, which can be obtained from natural
along with the coir fiber in S2, S3 and S4. It can resources and could offer a more sustainable
be concluded that S2 and S3 showed better solution. The outcome is expected to provide
tribological properties compared to other up-to-date solutions to the global transport
formulations. Hence, natural coir fiber can be industry and its friction material supply base
used as a candidate fiber or filler material for (Savage, 2007). Several research works have
the mass-scale fabrication of asbestos-free brake been carried out for the development of
pad without any harmful effect. asbestos-free brake pads (Dagwa and Ibhadode,
Keywords: Coir fiber, Natural fibre Composite 2006; Savage, 2007; Aigbodian et al., 2010).
(NFC), Friction coefficient, Wear, The use of bagasse (Aigbodian et al., 2010),
Microstructure. palm kernel shell (Dagwa and Ibhadode, 2006)
and hemp (Savage, 2007) have been
1. INTRODUCTION investigated in order to replace the asbestos-free
brake pad material.
Friction material is a heterogeneous material
and composed of a few elements and each No information is also available in literature on
material has its own functional properties such the use of coir fibre for the formulation of new
as to improve friction property at low and high brake pad material. Therefore, a new asbestos
temperature, increase strength and rigidity, free brake pad material has been developed with
prolong life, reduce porosity, and reduce noise the aim of using natural coir fibre as a
156
reinforcement or filler material in aluminium developed using powder metallurgy (PM)

matrix. Aluminum metal matrix composites (Al technique. This PM technique used in order to
MMCs) are attractive for their lightweight obtain uniform part and reducing tedious and
(three times lighter than cast iron) properties, expensive machining cost. The process begins
higher thermal conductivity, specific heat, with selection of raw material, weighing,
superior mechanical properties and higher wear mixing, compacting and sintering. Raw
resistance. Particulate reinforced aluminum materials were blended together in mini mixer
composite are promising candidate for to get evenly distributed ingredients. The pre-
automotive applications since they offer high form of the samples were fabricated using
specific stiffness and strength, good wear Automated Hydraulic Presses Machine to the
resistance and suitable thermal properties sizes of 25 mm x 25 mm x 6 mm. These pre-
(Telang et al., 2010). form samples were heated to 170°C and then the
compacted using 20 kg compression load with
Tribological behaviour of brake pad material 60 seconds holding time inside the compaction
plays an important role in deciding which die. After removing from compaction, the
formulation is suitable for brake system design for samples were sintered using automated oven at
a particular vehicle (Maleque et al., 2010). temperature of 200 0C for 5 hours. Finally, was
CHASE Dynamometer test is used in a ground using grinding machine according to the
laboratory scale in order to assess the size required for friction and wear test. Figure 1
tribological behavior of brake pad materials and shows the samples for tribological test.
to help in the screening of new material
formulations prior to dynamometer tests. The Table 1 : Formulation of Al-based natural fibre
test procedure using CHASE machine has been reinforced composite brake pad materials
described in Society of Automotive Engineer
standard SAE J661 (SAE J661, 1997). The edge MATERIALS Percentage (%)
code of the ‘normal’ and ‘hot’ friction values S1 S2 S3 S4 S5
has been described in SAE Recommended Aluminium
Practice J866 (SAE J866, 1996). The current powder 45 39 33 27 21
research attempts to examine the tribological Silicon
properties of Al-based natural fibre reinforced carbide 20 20 20 20 20
composite brake pad materials. Finally, the best Graphite 10 10 10 10 10
formulation was obtained based on the
characteristics performance of the candidate Coir fibre 0 5 10 15 20
formulations. Alumina 13 13 13 13 13
Zirconium
2. METHODOLOGY oxide 2 2 2 2 2
Paper ash 0 1 2 3 4
2.1 Materials
Phenolic
The raw materials in the form of powders and
resin 10 10 10 10 10
long fibre are used in the preparation of
laboratory sample for brake pad materials. The
ingredients contained filler, abrasive, solid
lubricant, binder, friction modifier and
additives. The coir fibre was used as a filler
material in this investigation which was
collected from waste coconut fruit and cleaned
thoroughly using ethanol to remove impurities.
It was crushed and ground to a fine powder
(with a range of 100-200 µm), and sieved using
sieve analyzer.
2.2 Sample Preparation

Five formulations such as S1, S2, S3, S4 and S5 Figure 1: Natural fibre reinforced (NFC)
which have different composition of coir fibre composite brake pad materials for tribological
contents (as shown in Table 1) were developed tests.
for friction and wear properties. However,
abrasive, solid lubricant, binder, friction
modifier and lubricants were kept same for all
formulations. The brake pad material was
157
2.3 Tribological Test during the application. Sample S2, S3 and S4

The NFC brake pad samples were showed higher friction coefficient with the
tested using a CHASE Dynamometer machine. value of 0.48, 0.54 and 0.55 respectively. The
The sample was fixed against a rotating brake friction coefficient decreases with increasing
drum with a constant rotating speed of 417 rpm rotating speed due to the formation of
under the load of 647 N in accordance with SAE graphite/carbon layer within matrix surface.
J661. Each sample was subjected to seven test
runs with the following sequences: (i) baseline, First Baseline
(ii) first fade, (iii) first recovery, (iv) wear, (v) 0.7
Friction coefficient (µ)

second fade, (vi) second recovery, and (viii) 0.6
baseline rerun. In the baseline run, the sample 0.5
0.4 S1
was subjected to an intermittent braking with 20 S2
applications, 10 seconds on and 20 second off. 0.3
S3
This was followed by continuous braking where 0.2 S4
0.1 S5
the temperature was allowed to rise from 82 ºC
to 288 ºC. In the first recovery run, the brake 0
0 10 20
was applied continuously but the drum Rotation per miute (rpm)
temperature is allowed to cool from 288 ºC to
93 ºC. The wear run was under intermittent Figure 2: The friction coefficient of the NFC
braking with 100 applications, 10 seconds on brake pad materials at first baseline condition.
and 20 second off while temperature was
maintained between 193 ºC to 216 ºC. First Fade
0.7
Subsequently, second fade and second recovery
0.6
run was performed similar to the first fade and
0.5
recovery except operated under higher
0.4 S1
temperature where it rises to 343 ºC. Finally,
0.3 S2
baseline rerun was performed similar to first test S3
0.2
condition. The weight and thickness of NFC S4
0.1
brake pad materials were taken before and after S5
0
the friction test. In order to obtain average
100 150 200 250
thickness value, three measurements were taken Temperature (ºC)
at different locations on the brake pad samples.
2.4 Morphological Analysis Figure 3: The friction coefficient of the NFC

The evolution of the surface morphology brake pad materials at first fade condition.
after the friction tests were studied by using
scanning electron microscopy (SEM) with an
EDX system. The surface of the samples was
examined with SEM in which an image is 0.7 First Recovery
created by secondary electrons ejected from the 0.6

surface. The EDX analysis was performed as a 0.5
spot and area using a primary energy of 15KeV. 0.4 S1
The samples were coated in order to obtain 0.3 S2
better image and micrograph on non-conductive 0.2 S3
S4
brake pad materials. 0.1
S5
0
3. RESULTS AND DISCUSSION 105 155 205 255
Temperature (ºC)
3.1 Friction from Link-Chase Dynamometer

Machine
Figure 4: The friction coefficient of the NFC
brake pad materials at first recovery condition.
The friction from CHASE dynamometer test is
shown in Figs 2 to 6 at different temperatures.
Figure 2 showed the samples run for first Figs 3 and 4 showed friction coefficient from
baseline. This condition represents the friction the first fade and recovery condition
level at 93 ºC as measured during 20 braking respectively for all samples. Fade refers to when
applications of 20s duration. The friction the friction material is subjected to successively
coefficient of S1 and S5 reveals constant profile elevated temperatures and recovery is when the
condition returns to the ambient temperature.
158
For the first fade condition, the temperature Table 2 shows that all the developed brake pad
(drum temperature) arises from 93 ºC to 288 ºC. materials exhibited lower friction coefficient
During the first fade, the S2 and S3 samples and average thickness loss. Even though friction
showed consistent in friction coefficient with coefficients of the developed formulations are
higher friction coefficient during fade lower than those of the commercial sample (0.5-
application with the value of 0.58 and 0.62 0.62µ (Talib, 2001)), but they are still higher
respectively. The rest of the samples did not than the minimum requirement of 0.15 as stated
show any trend of friction during the test run, in SAE J866 (1996). All the samples passed
but the friction coefficient decreases with during the second recovery test except sample
increasing the drum temperature. This S5. This is due to the lower value of friction
phenomenon is called fade, and the resistance to coefficient (0.026) in the minimum requirement
fade at high temperature is a critical requirement which is shall not be lower than 0.15. From the
for brake friction materials. table, sample S1 and S3 showed no change in
the class of friction coefficient either in normal
Figs 5 and 6 showed the friction behavior at or hot condition which is F-F. For sample S2 the
second fade and recovery conditions. For class changes from G to E, while for sample S4
second fade the drum temperature was increased the class also change from F at normal condition
to 343ºC. Sample S2 showed that the friction to E at hot temperature. Table 2 shows the
coefficient started to decrease at point of 150 ºC classification of friction coefficient of NFC
until meet the lowest friction coefficient of 0.18 brake pad materials. The values were taken
at temperature of 369 ºC. This trend also can be from second recovery condition.
discovered in sample S3, where the friction
coefficient of the sample decrease at point of Table 2: Classification of friction coefficient of
150 ºC until meet the lowest friction coefficient brake pad materials
of 0.21 at temperature of 369 ºC. As compared Samp Normal Hot friction Observati
to S2, friction coefficient of sample S3 is much le Friction ons/Rema
higher even at higher temperature with the value µ Co µ Code rk
of 0.21 at 369 ºC. For samples S1, S4 and S5, de
the friction coefficient decreased with S1 0.414 F 0.368 F Passed
increasing of drum temperature during the test S2 0.462 G 0.281 E Passed
run. S3 0.434 F 0.374 F Passed
S4 0.368 F 0.273 E Passed
0.7 Second Fade S5 0.274 E 0.158 D Failed
0.6
0.5 Figure 7 revealed the graph of friction
S1 coefficient which was obtained from wear test.
0.4 S2
0.3 S3 Sample S3 showed that after 20 cycles of
0.2 S4 braking application the friction coefficient of
S5 the sample started to decrease. But the trend is
0.1
0
really different from other samples. Sample S1
100 150 200 250 300 350 400 and S2 showed that the friction coefficient
Temperature (ºC) increased with increasing the cycles of braking
Figure 5: The friction coefficient of the NFC application. As in sample S1, there is not much
brake pad materials at second fade condition. change in friction coefficient during wear test.
Second Recovery
0.6
Friction behaviour during wear test
0.8
0.5
0.4 S1 0.6
S2
0.3 S1
S3 0.4
0.2 S4 S2
S5 S3
0.1 0.2
S4
0
0
100 150 200 250 300
Temperature (ºC) 0 10 20 30 40 50 60 70 80 90 100110
No of cycles
Figure 6: The friction coefficient of the NFC
brake pad materials at second recovery
condition. Figure 7: The friction coefficient of the
NFC brake pad materials during wear test.
159
3.2 Wear Properties of NFC Brake Pad general, the distribution of the ingredients was
Materials heterogeneous structure due to complex
Figure 8 shows the wear properties of NFC formation of new NFC material. Figure 9 (b)
brake pad materials of CHASE Dynamometer and (c) showed the secondary electron image of
wear test. The result shows the weight loss after the worn surface of the NFC brake pad
friction test. Sample S5 has the highest value of materials. In these figures, the direction of the
34.3 %, followed by sample S4 with its value of wear test is obvious especially on the darker
17.7 %. The sample with the lowest weight loss region indication that this region is softer than
was sample S1 with its value of 4.7 %. the other region. In Fig 9(c), there are some
Meanwhile, the thickness loss for sample S5 has porous area which were not abraded by the
the highest value which is 34.1 % because it is counterpart drum. The percentage of the
softer than others, followed by sample S4 with porosity in this material is approximately 10%.
value of 14.6 %. The sample with the lowest The surface topography is clearly revealed.
thickness loss is sample S1 with value of 4.5 %. However, some abraded regions are not
Analysis of test results shows sample S4 and S5 revealed in this figure. This is because
do not meet the requirements as stated in SAE secondary electrons provides the information
J866 (1996). Higher thickness loss means slightly below the worn surface not on the very
shorter brake pad life and this incurred more near surface. Bright regions in this figure
material and maintenance cost. Whereas, higher showed the element of higher atomic number
friction coefficient results in wheel locking at a and the darker region is for the element with
very much lower brake pedal force which could lower atomic number.
drastically change the vehicle braking Non contact
Abraded
characteristics. region
region
40
34.3 34.1
35 Contact
30 region
-softer region
25 Weight
% value
17.7 loss (%) (a) Sample S1 (b) Sample S2

20
14.1 14.6
15 12.3 Thicknes Silica
10.5 Abraded Aluminum
8.7 s loss (%)
10 region
4.74.5
5
0 Carbon
Porous
S1 S2 S3 S4 S5 area Coir
Sample fibre
Iron
Figure 8: Wear properties of NFC brake pad
(c) Sample S3 (d) Sample S4
material during wear test using CHASE
Dynamometer
Figure 9: SEM micrograph of worn surface (a)
3.3 Wear Morphology of NFC Brake Pad S1, (b) S2, (c) S3 and (d) S4 samples after Link-
Materials CHASE Dynamometer test.
The evolution of the surface morphology after

friction tests were studied by scanning electron
microscopy (SEM) and EDX analysis. After Table 3 shows the percentage (in weight) of the
interaction between the static partner and the compositions in each element analyzed by EDX.
sample, marks are usually left on the sliding EDX analysis shows the constituents present are
surface. These marks can convey useful aluminum oxide, iron, silicon and carbon. From
information about the contacting process. Each the observation, carbon is clearly shown by the
wear mechanism leaves its own features and the dark colour. Meanwhile, the less dark colour
mechanism can often be identified by SEM showed the compound that still dominated by
photomicrograph. Fig. 9 illustrates the the carbon. Sample S2 and S3 contained more
microstructure of NFC brake pad after wear carbon and less oxide. In this case, both SEM
testing using Link-CHASE dynamometer and EDX analysis show a very good agreement
machine. The grind section of brake pad was in order to demonstrate the morphology and
clearly visible with shiny metallic constituents surface chemical composition of the new NFC
and dark/grey non-metallic in the surface. In brake pad materials.
160
Using Bagasse. Tribology in Industry,

Table 3: EDX analysis of the NFC brake pad Volume 32, No. 1.12-18.
materials in weight% Dagwa, I. M, and Ibhadode, A.O.A. 2006.
Element S1 S2 S3 S4 Determination of Optimum Manufacturing
C 56.46 63.89 58.43 57.73 Conditions for Asbestos-free Brake Pad
O 34.58 24.64 33.53 34.05 using Taguchi Method, Nigerian Journal of
Engineering Research and Development,
Al 6.22 7.86 5.92 6.12
Basade Publishing Press Ondo, Nigeria Vol.
Si 2.66 3.50 2.01 1.84
5, No. 4, 1-8.
Zr 0.08 0.11 0.11 0.25 Savage. L., 2006. Eco-Friendly Brake Pads
Promise Greener Transport. Exeter
4. CONCLUSION Advanced Technologies, University Exeter.
Maleque M A, Dyuti S. and Rahman M M..
This research was focused on the tribological 2010. Material Selection Method in Design
study of Al–based natural fiber reinforced of Automotive Brake Disc. Journal of World
composites (NFC) brake pad materials using Congress on Engineering Vol III.
CHASE Dynamometer machine. Friction and Talib, R. J. 2001. A study on friction and wear
wear assessment tests were conducted on five characteristics for automtive friction
new friction material formulations with varying materials. Proceeding 6th Proceeding Asean
coir fiber contents from 0, 5, 10, 15, and 20 % Science and Technology Conference, 17 –
volume fraction along with binder, friction 19 September 2001. Bandar Seri Bangawan,
modifiers, abrasive material and solid lubricant. Brunei.
Based on tribological test results and Talib, R.J., Muchtar, A and Azhari, C.H. 2003.
morphological study on the worn surface, the Microstructural characteristics on the surface
following conclusions can be drawn: and subsurface of semi-metallic automotive
(a) The friction coefficient decreases with friction materials during braking process.
increasing rotating speed due to the formation of Journal of Material Processing Technology
graphite/carbon layer within matrix surface. 140. pp. 694-699.
(b) The weight and thickness loss increases with Telang A K, Rehman A ,Dixit G, Das S. 2010.
increasing temperature. Alternate Materials In Automobile Brake
(c) Sample S2 and S3 are showed better Disc Applications With Emphasis On Al
properties in terms of friction coefficient and Composites- A Technical Review. Journal of
wear. Engineering Research and Studies. pp. 35-
(d) The current coir natural fibre can be used as 46.
a candidate fibre or filler material for the mass- SAE J661, 1997. Brake lining quality test
scale fabrication of asbestos-free brake pad procedure, Society of American Engineers,
without any harmful effect. Warrendale, USA.
SAE J866, 1996, Friction coefficient
Based on the observations obtained from this identification system for brake lining,
study, it is possible to modify a specific Society of American Engineers, Warrendale,
tribological property of a brake friction material USA.
by changing the amount of coir fibre in a
systematic manner while expecting possible
changes in other tribological properties.
ACKNOWLEDGEMENT
This work was supported by FRGS research

grants scheme No. FRGS10-026-0145 from
Ministry of Higher Education, Malaysia.
Authors are grateful to the International Islamic
University (IIUM) and AMREC SIRIM which
made this study possible.
REFERENCES
Aigbodion, V. S., Akaide, U., Hassan, S.B.,

Asuke. F., Agunsoye J.O. 2010.
Development of Asbestos - Free Brake Pad
161
MODELLING ASPECT OF CORROSIVE WEAR UNDER BIODIESEL
A.A. Adebisi and M. A. Maleque
Department of Materials and Manufacturing Engineering

E-mail: debisi1@yahoo.com, maleque@iium.edu.my
ABSTRACT mechanism and their significant to the specific

environment. However, the quantification of the
Biodiesel as an alternative renewable fuel is corrosive wear synergism provides information on the
increasingly been considered as blending components required material for reducing total material loss for a
or replacement for conventional petroleum fuel. particular environment. Also corrosive wear could be
Although the acceptability of biodiesel in automotive minimized by expressing the wear degradation and
application is relatively a successful story, it is an issue corrosion in common terms which allows the
with regards to its materials incompatibility. When synergism between the degradation processes
compared to diesel, biodiesel is more prone to absorb qualitatively and quantitatively. Model development
water and allows the development of electrochemical for corrosive wear mechanism provides an insight into
corrosion. Corrosive wear occurs as a result of the behaviour of the synergism by controlling,
chemical reaction on wearing surface. The main aim of quantifying and optimizing the process. Madsen (1994)
this paper is to study the modelling aspect of corrosive developed a penetration rate equation model for
wear under biodiesel for automotive application. A quantifying each of the degradation process for both
systematic review analysis was performed on the corrosion and wear by providing useful information on
corrosive wear, wear of engine component materials the synergism between the processes. These models are
using on road engine test, aspect of corrosive wear and used to quantify the synergism result obtained from the
also factors affecting corrosive wear of automotive abrasive and sliding wear measurements. Batchelor and
component. A corrosive wear model has been Stachowiak (1988), Schumacher (1985); (1993) and
developed with three main domains such as mechanical Yue et al., (1987) have studied the synergism between
wear, electrochemical corrosion, and interaction of wear and corrosion. The main objective of this paper is
both processes to evaluate the total corrosive wear of to study the modelling aspect of corrosive wear under
automotive component materials under biodiesel. It biodiesel for automotive application by utilizing a
was found that modelling aspect play an important role systematic review analysis.
in the measurement of corrosive wear especially under
biodiesel. Different components of the automotive engine parts
such as filter, fuel liners, piston rings, fuel pump, fuel
Keywords: Modelling, corrosive wear, automotive injector, gaskets etc made from different materials such
materials, Biodiesel as aluminium, copper, brass, zinc, bronze, elastomers
have direct contact with fuel (biodiesel). Biodiesel
1. INTRODUCTION becomes more corrosive during usage or storage due to
degradation as a result of oxidation (Monyem, 2001),
Corrosive wear occurs as a result of chemical reaction moisture absorption (Thompson, 2007), attack by
resulting from the combination of a wear mechanism microorganism (Klofutar, 2007). Oxidation of
and a corrosive environment. The combined effect of biodiesel reconverts esters into different mono-
the two processes involves mechanical and carboxylic acids like formic acid, acetic acid, propionic
electrochemical mechanisms which often result to acid, caproic acid, etc which are responsible for
substantial increase in the rate of material loss and enhanced corrosion. This process also increases the
degradation [method] when compared to the individual free water content which is undesirable because it may
process of wear or corrosion action. The process of promote microbial growth and corrode fuel system
interaction among the operating mechanism such as components. Kaul et al, (2007) studied the corrosion
abrasion, impact, rubbing and corrosion etc influences characteristics for non edible oils like Jatropha curcas,
the total material losses and degradation especially in Pongamia glabra (Karanja), Madhuca indica (Mahua)
the presence of an aqueous environment. In a study and Salvadora oleoides (Pilu) using long duration
carried out by Dun (1985), corrosive wear has static immersion test for engine part like piston metal
seventeen synergistic relationships between abrasion, and piston liner. Biodiesel from salvadora showed
impact and corrosion that facilitate significant increase marked corrosion on the diesel engine compared to
to wear mechanism in wet and aqueous environments. neat diesel. Investigations carried out by Geller (2008)
The effect of these relationships can be minimized by showed that copper alloys are more prone to corrosion
acquiring fundamental knowledge of the interaction by biodiesel as compared with ferrous alloys.
162
The corrosion behavior of aluminum exposed to diesel engine passenger car in order to access the
biodiesel can be compared to the corrosion behavior of biodiesel impact on wear. The analysis of the
aluminum in aqueous solutions. It was also observed lubricating oil samples showed that the use of B50 and
that corrosion of aluminum can be used as a B100 may lead to increased wear in terms of higher
quantitative indication of biodiesel purity (Diaz- amounts of metallic elements, originated from the
Ballote, 2009). Maleque et al. (2000) and Kalam and different moving parts.
Masjuki (2002) observed that the wear rate in biodiesel
was relatively higher due to its oxidative and corrosive Investigations on wear in automobile engine for
behaviors. Corrosiveness and tribological properties of biodiesel derived from different sources like rapeseed,
biodiesel also depends on its feedstock (Bu inskas, palmoil, and soybean were carried out for both static
2009). Pure biodiesel sample from different origin has engine test and on-road engine test which involves
better value and lower tribological properties. static (fuel tank, filter, fuel pump injector housing, fuel
However, regular diesel fuel has more stable properties line, exhaust system, cylinder liner etc) and dynamic
compared to pure biodiesel as a result of difference in (piston, piston rings, inlet and exhaust valve, fuel
production process. pumps and filters plunger, connecting rod)
components. The dynamic components are mostly
2. WEAR ANALYSIS OF BIODIESEL metals and they slide against each other and also on the
static part during operation which results in the
Biodiesel is a realistic fuel alternative for diesel generation metal debris due to wear. Table 1
engines based on its lower energy content, higher summarizes the test conducted by analyzing the
cetane number, free of contaminants like sulphur and concentration of the metal debris in the lubricant oil
aromatics and most significantly burns cleaner than after running the engine for a particular period of time.
conventional diesel fuel. However, its capacity is been The wear rate, source of element and engine condition
limited as a result of feedstock availability. The need of provided relevant information from the test conducted.
biodiesel for automobile application raises concern on Biodiesel enhances better lubricity than diesel fuel as a
the impact of corrosion wear behaviour on automobile result of its inherent properties and the presence of
engine when it comes in contact with tribo-component components such as free fatty acids, monoglycerides;
of static and dynamic parts like piston, piston rings etc. diglycerides found in the biodiesel fuel which improves
The wear rate of materials in the presence of biodiesel the lubricity of the fuel (Knothe, 2005). Oxygen
occurs in various forms which amount to the evaluation containing compounds such as free fatty acids, esters
of metal debris generated as a result of wear. Fontaras are superior wear and friction reducing agents (Haseeb
et al., (2009) investigated the metal particulate et al., 2010). These compounds adsorb or react on
concentration in neat soybean oil biodiesel (B100) and rubbing surfaces to reduce adhesion between
its 50 vol% blend with petroleum diesel (B50) on contacting asperities and thereby limit friction, wear
and seizure.
Table 1 elemental analysis for on-road engine test results on wear in biodiesel as compared with diesel
Wear elements
Sources Biodiesel Engine Operation Hour References
Al Cr Cu Fe Pb Zn
Ethyl soyate B100 200 H L H L H - Clark et al. (1984)
Methyl soyate B100 200 L S L L H - Clark et al. (1984)
Rapeseed B20 512 L L L L L - Agarwal et al. (2003)
Rapeseed B100 1000 L L - L L - Perkins et al. (1991)
Rapeseed B50 1000 L L - L L - Perkins et al. (1991)
- Prateepchaikul and
Palm oil B100 1000 H H H S H
Apichato (2003)
Palm oil B7.5 100 L - L L L - Kalam et al., (2002)
Palm oil B15 100 L - L L L - Kalam et al., (2002)
Palm oil B7.5 300 - - L L L H Hu J. et al. (2005)
Linseed oil B20 500 - L L L L L Agarwal et al. (2003)
L = less wear; S = similar wear; H = higher wear
3. CORROSION OF BIODIESEL free fatty acid which may exist as a consequence of
incomplete transesterification reaction and the ability
Biodiesel is a vegetable based oil or animal fat- of the biodiesel to absorb moisture (hygroscopic
based diesel fuel consisting of saturated and nature) increase the corrosiveness of the fuel.
unsaturated long chain fatty Moreover, the auto-oxidation of biodiesel can also
acid alkyl (methyl, propyl or ethyl) esters and are catalyze the corrosion characteristics.
produced chemically by reacting lipids e.g., vegetable Transesterification (also called alcoholysis) reaction of
oil, animal fat (tallow) with alcohol. The presence of the fat or oil triglyceride with an alcohol forms esters
and glycerol. Figure 1 shows the transesterification
163
reaction of triglycerides. A catalyst is usually used to This process also increases the free water content
improve the reaction rate and yield. Because the which is undesirable because it may promote microbial
reaction is reversible, excess alcohol is used to shift the growth and corrode fuel system components. Several
equilibrium to the product side. The reaction results in investigation have been carried out to determine the
a compound called fatty acid alkyl ester (biodiesel) and corrosiveness of different biodiesel and the corrosion
a by product, glycerol. From the resulting reaction, effect on material compatibility for automotive
hygroscopic biodiesel may be formed due to component materials like aluminium, copper, stainless
persistence of mono and di glycerides left over from an steel and leaded bronze using static immersion test
incomplete reaction process. Biodiesel degrades (Fazal et al., 2010 and Haseeb et al., 2010) In
through the process of oxidation, moisture absorption automobile fuel system application, components made
and attack by microorganism during storage or use. from aluminium are piston (100%), engine block
The oxidation of biodiesel reconverts esters into (19%), cylinder head (70%) etc. Other components
different mono-carboxylic acids like formic acid, acetic made from stainless steel include valve bodies, fuel
acid, propionic acid, caproic acid, etc which are filter, nozzle and pump ring.
responsible for enhanced corrosion (Tsuchiya, 2006).
Fig. 1 Transesterification reaction of triglycerides
Moreover, copper and its alloy are used to manufacture per year, W is the weight loss (mg), D is the density
injector, pump and bearing (Fraer, 2005). All these (g/cm3), A is the exposed surface area (square inch)
component parts made from different materials come in and T is the exposure time (h). Haseeb et al. (2010)
contact with fuel which undergoes chemical reactions investigated the corrosion behaviour of commercial
and thereby deteriorate the material as well as degrade pure copper and lead bronze in automotive fuel system
the fuel properties too. Kaul et al., (2007) investigated for static immersion test in B0, B50, B100 at room
the corrosiveness of different biodiesel (i.e. Jatropha temperature for 2640hrs and B0, B100, B100
curcas, Karanja, Mahua and Salvadora) as compared to (oxidized) at 60°C for 840hrs. The result shows that
that of diesel fuel. They found that biodiesel from pure copper was more susceptible to corrosion in
Jatropha curcas and Salvadora were more aggressive biodiesel compared to leaded bronze. In another study,
for both ferrous and non-ferrous metal. Geller et al., Fazal et al, (2010) studied the corrosion comparison of
(2008) observed that copper alloys are more prone to aluminium, copper and stainless steel in both diesel
be attracted by corrosion in fat based biodiesel as (B0) and biodiesel (B100) using immersion test at
compared with ferrous alloys. It was also reported that 80°C for 1200hrs. They observed that the effect of
pitting corrosion was found on sintered bronze filters in corrosion and change in fuel properties upon exposure
oil nozzle after 10 h of operation with biodiesel at 70 to metal is more in biodiesel than diesel. They
°C (Sgroi et al., 2005). In another study, corrosion concluded that copper and aluminium were susceptible
attack was also reported even for lower biodiesel (2%) to attack by biodiesel whereas stainless steel was not.
blend levels (Tsuchiya, 2006). The corrosive and
oxidative nature of biodiesel relatively increases wear 4. FACTORS AFFECTING CORROSIVE WEAR
rate (Haseeb et al., 2010).
4.1 Materials properties
Research conducted to determine the extent of
corrosion on automotive component materials utilized The properties of all the materials involved in the
weight loss measurement and change in surface tribological contact and the reaction products formed
morphology (Fazal et al., 2010). The data obtained on the rubbing surface and surrounding environment
from the weight loss were analyzed and converted into are of relevance to determine the material behaviour. In
corrosion rate by using the equation; the absence of corrosion, wear resistance of a material
depends on properties such as hardness, rigidity,
w  534 ductility and yield strength. The relationship between
Corrosion rate (mpy )  these properties on corrosive wear rate requires an in
D T  A (1) depth study for better understanding of the synergistic
mechanism. Published research has tried to study the
where corrosion rate ‘mpy’ stands for mils (0.001 in.) synergistic effects between wear and corrosion
164
processes which result in accelerated material loss and • Normal force

in some cases actually decelerate material loss. • Sliding velocity
• Type of motion
4.2 Mechanical operation • Shape and size of
Mechanical contacting body
• Alignment
The rate of corrosive wear for a given metal • Vibration
environment combination depends on the applied
forces and the type of contact sliding, fretting, rolling Electrochemical Corrosive wear
Environment
or impact. The other factors include sliding velocity, systems
type of motion, shape and size of contacting bodies, • Applied potential • Viscosity
alignment, vibration etc. For example, in the case of • Ohmic resistance • Conductivity
• Film growth • pH
fretting corrosion, there are small amplitude Materials
• Corrosivity
• Repassivation kinetics
oscillations occurring in a corrosive environment. • Temperature
• Valence • Hardness, Plasticity
Contact geometry involving shape and size of • Active dissolution • Surface roughness • Dissolved oxygen
contacting surfaces is another important parameter in • Microstructure, inclusions • Air
corrosive wear, as it determines the size of the contact • Oxide film properties • Humidity
zone and the alignment of the rubbing surfaces. • Materials transfer
Different mechanical processes have different • Material transfer
parameters affecting the process, like in case of erosion
the energy and the angle of incidence of the impacting
particles and their shape are critical variables. Fig. 2 Factors influencing corrosive wear
4.3 Solution and environment

5. CORROSIVE WEAR MODEL
The environmental variables/parameters and contact
conditions (metal pair or non-metal pair), play an 5.1 Corrosive wear interactions
important role in corrosive wear. Its influence is in the
form of the medium at the interface i.e., solid, liquid or Corrosive wear interactions can be expressed as wear-
gaseous and its corresponding properties like viscosity, corrosion, where the total damage, T, is given by (i.e.
conductivity, pH, corrosivity, temperature etc. For sliding or abrasive wear):
example, the metals exposed to air, the relative
humidity will determine whether a thin liquid DT = WM + CS + SI (2)
electrolyte film may form at the surface changing the
corrosion mechanism. In case of aqueous systems, where, WM is the mass loss due to wear, CS is the
concentration of oxygen, pH and concentration of solids free flow corrosion rate and SI is the synergistic
certain anions like chloride ions influences corrosivity. or interactions term. Total damage under erosion–
In case of high temperature applications, the physical corrosion, DT, can be represented as:
nature of the scales formed is critical. Certain corrosion
products containing sulphur have a relatively low DT = CM + CS + SI (3)
melting point, which can lead to the formation of
highly corrosive molten salts on the surface. where CM is the corrosive wear material loss,
There are numerous interactions between wear and
4.4 Electrochemical parameters corrosion and many ways to experimentally quantify
them. Mostly these interactions are referred to as
In corrosive wear system, the corrosion monitoring is synergistic (two or more parameters interacts together
analyzed by using basic electrochemistry process. to produce a result) and as attain the level of synergy.
Basic parameters applied are potential, ohmic Synergy is the difference between wear-corrosion and
resistance, passive film growth, active dissolution etc, the summation of its two parts and can be expressed by
as shown in Fig. 2. The electrochemical aspect is equation (4).
considered mainly because corrosive wear phenomena
have been studied for many years by electrochemists SI = DT − (CM + CS) = (ΔCW + ΔMW) (4)
and tribologists. Electrochemists have concentrated
their attention on the study of kinetics of repassivation where DT, CS and CW are typically gravimetric terms
of metal surfaces activated by scratching, whereas relating to wear-corrosion, electrochemical corrosion
tribologists have been interested as to how surface and mechanical wear mechanisms, respectively. The
oxidation during rubbing affects the rate of mechanical interactive processes can be simplified into two
wear. components, ΔCW and ΔMW, where ΔCW is the
corrosion-enhanced wear and ΔMW is the
mechanical/wear-enhanced corrosion. Recent literature
has defined ΔCW as the synergy term and ΔMW as the
additive term.
165
The synergistic effect (interactive term), SE, is referred reliable result. Figure 3 shows the proposed model
to as ΔCW or (ΔMW + ΔCW) depending on the literature which consist of a combined system, the tribometer to
source and under what conditions CS has been measure evolution of wear depth (linear wear) and the
obtained. Synergistic levels for different materials must electrochemical set up to monitor the corrosion rate
be carefully understood when using multiple sources of from the model system. The electrochemical set up
literature. The ASTM G119-93 standard is a very consists of three electrodes as indicated by (c), (e) and
useful guide to measure and evaluate synergy (g). The sample is operating as the working electrode.
(Ponthiaux et al., 2004). The proposed model clearly shows that the corrosion
phenomenon and the mechanical wear of the material
5.2 Proposed Model for corrosive wear under or in the presence of biodiesel can be measured
simultaneously or concurrently and therefore provide
The corrosive wear laboratory setup model needs to be the information on corrosive wear.
properly configured in other to provide an accurate and
Fn (load)
a
F, D Key features
b (a) applied load
(b) load cell
g (c) counter electrode
h (d) biodiesel
c
i (e) working electrode
d (f) speed controller
(g) reference electrode
e j
(h) voltmeter
(i) data acquisition system
(j) ammeter
(k) motor
f k
Fig 3 Proposed model for corrosive wear study of biodiesel

Figs. 4 illustrate sectioned profile of corrosive wear in
From the above it can be said that corrosive wear presence of biodiesel. Relative to a non- corroded
consist of three main domain, wear due to mechanical region on a datum surface, the total corrosive wear can
action, wear due to corrosion and wear due to the be expressed as;
interaction of both corrosion and mechanical action
under synergistic conditions. WCW = CW + MW + WIN (5)
Where,
CW WIN = ΔMW + ΔCW (6)
Therefore,
MW WCW
WCW = CW + MW + (ΔMW + ΔCW) (7)
Where WCW = total corrosive wear

WIN CW = wear due to corrosion
MW = wear due to mechanical action
Unaffected region (Non corroded region) WIN = wear due to corrosion and mechanical
action
Fig 4 Model of sectioned profile of corrosive wear It is expected that the total corrosive wear, w cw can be
measured using equation 7 and prediction can be made
166
for the safe use of automotive component materials biodiesel." Electrochemistry Communications
under biodiesel. 11(1): 41-44.
6. CONCLUSION Dunn, D. (1985). "Metal removal mechanisms

comprising wear in mineral processing." Wear
Based on the current study along with the proposed of Materials 1985: 501-508.
corrosive wear model the following conclusion can be
drawn from the analysis; Fazal, M., A. Haseeb, et al. (2010). "Comparative
corrosive characteristics of petroleum diesel and
i. The current study has been able to provide a palm biodiesel for automotive materials." Fuel
general knowledge about the field of corrosive Processing Technology 91(10): 1308-1315.
wear.
Fontaras et al., T. T., L. Ntziachristos, and Z. Samaras
ii. Corrosive wear system is significantly (2009). Biodiesel (soy-bean FAME) effect on
influenced by factors such as mechanical particulate and gaseous pollutants from a
action, material properties and the passenger car. European Aerosol Conference.
environmental condition. Karlsruhe, Germany: 4.
iii. It is established from the study that corrosive Fraer, R., H. Dinh, et al. (2005). "Operating Experience
wear mechanism has three domains such as and Teardown Analysis for Engines Operated
wear due to mechanical action, wear due to on Biodiesel Blends (B20)." SAE Technical
corrosion and wear due to the interaction of Paper: 01-3641.
both corrosion and mechanical action under
synergistic conditions. Geller, D. P., T. T. Adams, et al. (2008). "Storage
stability of poultry fat and diesel fuel mixtures:
iv. The successful development of the model specific gravity and viscosity." Fuel 87(1): 92-
shows that the total corrosive wear can be 102.
obtained from the model set up.
Haseeb, A. S. M. A., H. H. Masjuki, et al. (2010).
"Corrosion characteristics of copper and leaded
Acknowledgement bronze in palm biodiesel." Fuel Processing
Technology 91(3): 329-334.
This work was supported by FRGS research grants
scheme No. FRGS10-026-0145 from Ministry of Haseeb, A., S. Sia, et al. (2010). "Effect of temperature
Higher Education, Malaysia. Authors are grateful to on tribological properties of palm biodiesel."
the International Islamic University Malaysia (IIUM) Energy 35(3): 1460-1464.
which made this study possible.
Hu, J., Z. Du, et al. (2005). "Study on the lubrication
REFERENCES properties of biodiesel as fuel lubricity
enhancers." Fuel 84(12-13): 1601-1606.
Agarwal, A., J. Bijwe, et al. (2003). "Wear assessment
in a biodiesel fueled compression ignition Kalam, M. and H. Masjuki (2002). "Biodiesel from
engine." Journal of engineering for gas turbines palmoil--an analysis of its properties and
and power 125: 820. potential." Biomass and Bioenergy 23(6): 471-
479.
Batchelor, A. and G. Stachowiak (1988). "Predicting
synergism between corrosion and abrasive Kaul, S., R. Saxena, et al. (2007). "Corrosion behavior
wear." Wear 123(3): 281-291. of biodiesel from seed oils of Indian origin on
diesel engine parts." Fuel Processing
Bu inskas, V. and R. Suba ius (2009). "Tribological Technology 88(3): 303-307.
Properties of Biodiesel Fuel and Its Mixtures
with Diesel Fuel." Klofutar, B. and J. Golob (2007). "Microorganisms in
diesel and in biodiesel fuels." Acta chimica
Clark, S., L. Wagner, et al. (1984). "Methyl and ethyl slovenica 54(4): 744-748.
soybean esters as renewable fuels for diesel
engines." Journal of the American Oil Chemists' Knothe, G. and K. R. Steidley (2005). "Lubricity of
Society 61(10): 1632-1638. components of biodiesel and petrodiesel. The
origin of biodiesel lubricity." Energy & Fuels
Díaz-Ballote, L., J. López-Sansores, et al. (2009). 19(3): 1192-1200.
"Corrosion behavior of aluminum exposed to a Madsen, B. "Standard guide for determining amount of
synergism between wear and corrosion." ASTM
167
G119-93, 1994 Ann. Book ASTM Stand 3: 507- Schumacher, W. (1985). "Corrosive wear synergy of
512. alloy and stainless steel." Wear of Materials
1985: 558-566.
Maleque, M., H. Masjuki, et al. (2000). "Effect of
mechanical factors on tribological properties of Schumacher, W. (1993). "Corrosive wear principles."
palm oil methyl ester blended lubricant." Wear Materials Performance;(United States) 32(7).
239(1): 117-125.
Sgroi, M., G. Bollito, et al. (2005). "BIOFEAT:
Monyem, A. (2001). "The effect of biodiesel oxidation Biodiesel fuel processor for a vehicle fuel cell
on engine performance and emissions." Biomass auxiliary power unit:: Study of the feed system."
and Bioenergy 20(4): 317-325. Journal of power sources 149: 8-14.
Perkins, L., C. Peterson, et al. (1991). Durability Thompson, et al. (2007). "Moisture absorption in
testing of transesterified winter rape oil biodiesel and its petro-diesel blends." Applied
(Brassica napus L.) as fuel in small bore, multi- Engineering in Agriculture 23(1): 71-76.
cylinder, DI, CI engines, Society of Automotive
Engineers. Tsuchiya, T., H. Shiotani, et al. (2006). "Japanese
Standards for Diesel Fuel Containing 5%
Ponthiaux, P., F. Wenger, et al. (2004). FAME: Investigation of Acid Generation in
"Electrochemical techniques for studying FAME Blended Diesel Fuels and Its Impact on
tribocorrosion processes." Wear 256(5): 459- Corrosion." SAE Technical Paper: 01-3303.
468.
Yue, Z., P. Zhou, et al. (1987). "Some Factors
Prateepchaikul, G. and T. Apichato (2003). "Palm oil Influencing Corrosion--Erosion Performance of
as a fuel for agricultural diesel engines: Materials." Wear of Materials 1987. 2: 763-768.
comparative testing against diesel oil."
Songklanakarin Journal of Science and
Technology 25(3).
168
EXPERIMENTAL EVALUATION OF INDUSTRIAL RUBBER WEAR USING AKRON AND

TRIBOMACHINE
R. M. Nasir*, A. S. Othman, A. Y. Saad and N. Othmanb

School of Mechanical Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Seberang Perai Selatan, Pulau
Pinang, Malaysia.
b
School of Materials & Mineral Resources Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Seberang Perai
Selatan, Pulau Pinang, Malaysia.
Corresponding Author Email: meramdziah@eng.usm.my*
ABSTRACT
The critical mechanical and tribological The tire traction as extensively investigated by
aspects of industrial tire were carried out including Heinrich et al.(2007) emphasized on velocity and load
tensile, elongation at break, tensile modulus, hardness, depending on friction coefficient. According to Persson
rheology, and abrasion tests. The masticating process (2001) rubber has low elastic modulus, rubber often
for formulated rubber was done by two roll mill exhibits elastic instabilities during sliding. Dick (2000)
machine using conventional vulcanization method. The added that natural rubber has high resilience and low
abrasion test was carried out using both Akron and heat buildup during flexing because of its hysteretic
Tribometer to evaluate the volume loss per revolution. properties. Hanafi et al. (1998) also supported that
Eventually, the worn surface was examined using natural rubber has low compression, set and stress
Scanning Electron Spectroscopy (SEM) to give a relaxation, good electrical insulation and good
reasonable vision of surface deformation. resistance to abrasion, tear and fatigue.
Keywords: tire, wear, masticating, dry abrasion, SEM There are thousands of proprietary secret recipe used in
rubber industry. Hence, the price of rubber has been
accelerating to ceiling price due to its many
applications and its specifications. Last but not least,
1. INTRODUCTION performance: The tire performance is highly influenced
by crown contour contact with road surfaces or ground
Past researchers (Stack,2009, Bhowmick et al.,1994,
to optimize the durability, cornering force, rolling
Freakly et al., 1998 Blow et al. 1988 and
resistance and fuel consumption according to Cho et
Lindley,1992), has developed specialties rubber with
al.(2004). For this project, Standard Thailand Rubber
low rolling, high swerving resistance, high and stabile
(STR) 5L as natural rubber (NR) was masticated with
mobility, low road noise, comfortable and safe driving
styrene butyl rubber (SBR) as synthetic rubber (SR).
under broad cyclic operation are critical factors for
Polymers containing a high proportion of styrene are
tires competitiveness. In rubber industries, certain type
widely used in blends with diene rubbers to improve
embedded fibers, steels and textiles for cords in belt,
wear resistance and high hardness with low specific
bead, and ply parts has been employed as described by
gravity.
Gent (2000).
Hofman (2000), Braun (2003) and Cho et al. (2004) 2. EXPERIMENTAL DETAILS
have explained the rubber technology that lead to
adhesion, hysteresis and cohesion loss friction which A-The Recipe
generate intermolecular force, deformation and rupture
in rubber tread respectively due to rough, uneven and The formulation is the heart of rubber compounding.
sharpened road surface. The summary of skid For this project, a rubber recipe for a tire is shown in
resistances i.e. in dry, wet, snow, and ice are Table 2. This recipe is based on two raw rubbers which
contributed by different degrees of frictional forces are NR and SR, reinforcing filler used is CB, an anti-
which are shown in Table 1. degradants package (TMQ and IPPD), and a curing
system (zinc oxide, stearic acid, sulphur, and CBS).
B-Manufacturing Process
169
For the manufacturing process; when non-rubber materials are added to the rubber by a certain mixing
ingredients is added and mixed into uncured rubber cycle. Table 3 shows a typical mixing cycle which is
compound in internal mixer; viscoelastic fluid will be been followed while using the two roll mill. To prevent
formed. Rheological tests were conducted before the scorching phenomenon, the vulcanizing agent
molding processes at high pressure and temperature. (sulphur) and accelerator typically is added at the end
The curing time obtained from the rheology test will be of the mixing cycle. Roller surface temperature can be
used in moulding test press. Thus, the quality, controlled using the water cycle which is hot or cold
marching, plateau, or reversion cured of rubber water. Thickness of rubber compounds produced
compound produced could be determined. The roller depends on the size of the opening between the two
speed ration between back rollers to front roller is in rollers as shown in Table 4.
the range 1.1-1.5:1. The compounding ingredients such
as filler, accelerator, activator, antioxidant, and other
Table 1: Skid Resistance (Dick, 2000)
Adhesion Hysteresis Cohesion

Table Skid Resistance Degree of Effects 2:
Rubber (safety factor, f) Recipe
Dry (f – 0.6) High Med Low
Wet (f – 0.3 – 0.6) Med High Low
Snow (f – 0.3 – 0.6) Between wet to ice condition
Ice (f – 0.3) Middle Middle Low
Ingredients phr Weight (g) Compound function
STR 5L, Natural Rubber 75 157.89 Rubber

SBR, Synthetic Rubber 25 52.63 Rubber
N330 Carbon Black 30 63.16 Reinforcing Agent
IPPD 2 4.21 Anti-degradant
TMQ 2 4.21 Anti-degradant
Stearic Acid 2 4.21 Activator
Zinc Oxide 3 6.32 Activator
CBS 1.5 3.16 Accelerator
Sulphur 2 4.21 Vulcanizing Agent
Table 3: Typical Compounding Cycle for Two Roll Mill

Step Compounding Cycle Time (minute)
1 A small friction of rubber is passed through the
mill without stripping for two or three times. 1
2 The rubber is stripped with size opening of 0.03
inch until the rubber surface is smooth. 4
3 Zinc oxide is added uniformly across the mill.
When the zinc oxide was compounded, make a
three times cutting to the rubber from each side
so that the uniformly distributed of zinc oxide
can be produced. 8
4 Add the stearic acid. 2
5 Add all the compounds ingredients except that
the accelerator and the vulcanizing agent at the
end of the mixing cycle. 4
6 Make a three times cutting to the rubber same
in the third step. 2
7 Cut the rubber from the mill and the rubber is
passed through the mill for six to seven times at
opening size of 0.1 inch. Then, weight the rubber. 3
170
Table 4: The Relationship between Roller and Rubber – 8 MN/m2 is given while compressing. The mould has
Size been placed between two plates. For this experiment,
Roller Size (Inch) – The Rubber Size (kg) the vulcanizing temperature used is 150oC. The plates
Diameter x Length and mould will be heated by saturated steam, hot oil, or
6 x 13 0.6 – 0.9 electricity. After the mould is closed, the rubber
10 x 20 2.2 – 3.6 compound was placed between the metal mould.
28 x 84 28 - 35 Usually, this method is used to prepare a vulcanized
rubber for testing of tensile strength, elongation at
For this project, the compression moulding was used break, and modulus, 2mm thickness was fabricated.
because it is simple. A constant hydraulic pressure of 4 For abrasion resistance and hardness tests 13mm
thickness was molded as shown in Fig.1 .
(a) (b) (c) (d)
Figure 1 (a) Slap with 5 mm thickness(b) slap with 2 mm thickness (c) disc with 13 mm thickness (d) dumbell shape
specimen
C-Mechanical Tests using the tensile machine (Instron 3366) with cross-
head speed 50 cm/min.
Tensile properties are determined by stretching
standard test pieces at a constant rate until it breaks Hardness tests method can be obtained from ISO 48 for
using Universal Tensile machine. The method can be rubber sample in the range of 30-85 IRHD
obtained in standard ISO 37, BS 903 and ASTM 0412. (International Rubber Hardness Degrees). The hardness
For the test, the specimen used is dumbbell shape tester Techlock GS – 706G was used with average
(2mm thick) as shown in Fig. 1(d). The tensile hardness value
properties of vulcanized rubber usually determine
(a) (b)
Figure 2 (a) Akron machine (b) Tribotester set up with rubber specimen
Abrasion is done by producing relative motion The differences between Akron and Tribomachine is
between the rubber and an abrasive surface (abradant) that for Akron, the rotating part is the rubber and the
pressed together by a predetermined force. Abrasion abradant but for tribotester, only the abradant is
loss is generally measured the volume abraded per unit rotating and the rubber is in static position. Besides, the
of energy or revolutions expended indeed loss per unit abradant used is also different for both machines. For
energy is used in some other tests. The Akron abrasion Akron, an abrasive wheel is used but for the tribotester,
tester (as shown in Fig 2 (a)) is most widely used. In an abrasive paper was used as the abradant.
order to determine a securing specific weight of the
rubber, a small piece of the specimen was cut with a For rheological test, a Monsanto MDR 2000 machine
knife. Then, Precisa weight machine was used to weigh was used. The sample used for the test is 2.8 g in
the specimen. weight. The heat transfer and balance terms can
achieved in a short time. Cavity and sample
171
temperature can be changed quickly because this Table 5 the average tensile strength is 23.39 MPa,
machine did not have a rotor. Torque value is usually elongation at break at 766.7% which is within
given in units of kgf.cm or dNm. industrial target. The modulus for both at 100% and
300% extension is at 1.566 and 5.720 MPa is also
D-Optical test within the industrial target.
Scanning electron microscope (SEM) namely

B-Hardness Test
Hitachi Table Microscope TM-1000 with high-
The average hardness for the specimen is 56.67
resolution and narrow electron beam, with new
(IRHD) which is comparable with the industrial rubber.
graphical user interface was used
C-Abrasion Test
3. RESULTS AND DISCUSSIONS Density of the rubber compound = 1.072 kg/m3
A-Tensile Test
The average thickness for all samples is between 1.82

to 1.87 mm and consistent width of 6.4 mm. From
Table 5: Tensile Properties
Sample Tensile Strength Elongation at Modulus at 100% Modulus at 300%
(MPa) break (%) (MPa) (MPa)
1 23.10 816.7 1.515 5.308

2 24.70 781.7 1.583 5.726
3 22.37 701.7 1.599 6.127
Average 23.39 766.7 1.566 5.720
SD 1.18 58.9 0.045 0.410
Table 6: Akron machine

Specimen Time Total Abradant Initial Final ΔW Volume Volume loss/rev
(Minute) Revolutions(rp mass (g) mass (g) loss (m3)
m) (g)
1 5 536 39.35 39.26 0.09 0.0840 1.566x10-4
2 10 529 39.29 39.09 0.20 0.1866 1.752x10-4
3 15 530 39.09 38.77 0.35 0.3265 2.047x10-4
4 20 510 38.71 38.26 0.45 0.4197 1.994x10-4
5 25 513 38.26 37.68 0.58 0.5410 2.067x10-4

6 30 501 37.68 37.01 0.67 0.6250 2.004x10-4
Table 7: Tribomachine
Specimen Time Total Abradant Initial Final ΔW (g) Volume Volume
(Minute) Revolutions(rp mass (g) mass (g) loss (m3) loss/rev
m)
1 5 7250 3.2139 3.1071 0.1068 0.0996 1.374x10-5
2 10 7250 3.2277 3.0109 0.2168 0.2011 1.395x10-5
3 15 7250 3.2306 2.8086 0.4220 0.3937 1.809x10-5
4 20 7250 3.2543 2.7061 0.5482 0.5114 1.763x10-5
172
5 25 7250 3.2599 2.4063 0.8536 0.7963 2.197x10-5
6 30 7250 3.2934 2.2795 1.0139 0.9458 2.174x10-5
Experimental Evaluation between Akron and Specific Wear Rate at different revolutions
Tribomachine Total revolutions (rpm)
7250 14500 21750 29000 36250 43500
0.0014 0.00014
Specific volume loss per rev.

The abrasion test has been carried out with two 0.0012 0.00012
different machines which are Akron abrasion tester and 0.001 0.0001
(m3/rpm)
Tribomachine as shown in Table 6 and 7 respectively. 0.0008 0.00008
0.0006 0.00006
From the results shown in Figs. 3 and 4, it can be seen
0.0004 0.00004
that the Akron abrasion tester gave a better result than 0.0002 0.00002
the tribomachine because Akron has a lower wear rate 0 0
536 1065 1595 2105 2618 3119 Akron machine
than the tribomachine. Although tribomachine has Tribomachine
Total revolutions (rpm)
higher wear rate but it is because that the abradant
rotates longer with different set up than the Akron
machine. From the results, it can also be seen that the Figure 4 Specific wear rate at different revolutions for
volume loss/rev for tribomachine is lower than the Akron and tribomachine
Akron machine and it is fluctuating in increasing
pattern for both experiments. Thus, both methods are
comparable.
Volume loss for STR blend SBR
1
Volume loss (m3)
0.8
0.6 Akron machine

0.4 Tribomachine
0.2
0
5 10 15 20 25 30
abrading time (min)
Figure 3 Volume loss against abrading time for Akron

and tribomachine
D-Rheology Test
Relationship of rheological test with curing characteristics
173
Induction Curing Past due

Stage Stage Stage
0.9(Tmax-Tmin)
Tmax
Tmin + 2
Tmin Figure 5 Graph of Cure Rate for the rubber compound
The induction stage At the maturation stage, the vulcanizing agent

was used until it decreased or finished and cured
At this stage, the torque will go through until past due stage. At this stage, there are three
the minimum phase (Tmin). At Tmin or before, the possibilities that will happen where the torque
cross linking not yet happen in the rubber will increase (marching cured), consistent
compound. After Tmin , the torque began to (plateau cured), or decrease (reversion cured).
increase showing that the cross linking occurred. The behavior that will experienced by a rubber
The time taken from the initial measurement (t=0) compound depend on the type of rubber, type and
to reach T=Tmin+2 is scorch time. Scorch time is amount of accelerator, activator, and vulcanizing
also referred as induction time. Practically, scorch agent or in other words, the vulcanization system
refers to the temperature time which depends on used. For this research, the conventional
heat processing before there are signs that vulcanization was used because the amount of
crosslink was started. Scorching is a premature sulphur is more than the amount of accelerator.
phenomenon and should be avoided while This is shown in Fig. 5.
forming because rubber will be quite hard to
process. This is shown in Fig. 5. E-Optical test
The curing stage The SEM images of unworn rubber

specimen is shown in Fig. 6. The image after the
After the induction stage, torque will test is shown in Figs. 7-10. Fig. 7 shows some
increase rapidly to show that the maturation has ridges pattern formation known as Schallamach
occurred to the rubber compound. The increasing waves .
degree of maturation or the degree of crosslink
measured by the increased in torque required to Now, the problem is that there is
maintain certain amplitude of testing. The needle/fiber structure that can be seen from the
amplitude used is small and the disc moves in structure which is shown in Fig. 8. It is obvious
reciprocal and non-continuous, then the crosslink that there is a lot of debris in the compound
formed does not broken. In this stage also, the although the ingredients from the formulation
rubber compound is firm and elastic. So, the totally not having a single sample that have the
torque is a measure of modulus. It means that the needle/fiber structure. However, the needle/fiber
torque will increase with the increasing degree of structure can be seen after the test and not before
crosslink. The increasing of torque while the test as shown in Fig. 9 compared to Fig. 6.
maturation is considered proportional to the Therefore, a cryo fracture test has been carried
number of crosslink formed per unit volume of out in order to prove that whether the compound
rubber. This is shown in Fig. 5. contains this kind of structure or not.
The past due stage

174
The cryo test is a test to see the

dispersion of the structure in the sample. The
sample will be dip into the liquid nitrogen for 5 –
10 minutes. Then, the sample will be break to two
pieces and after that the pieces will taken for
SEM test. So, the image after the cryo test is
shown in Fig. 10. From the image, the
needle/fiber structure is actually not from the
rubber compound but it comes from somewhere
else. It is comes whether from the brush, samples
preparation, or from the machine itself because
the structure is not embedded to the compound.
Needle/fiber
Thus, it is because of impurities and does not
affect the results obtained. structure
Figure 8 The image of debris formed in the rubber
compound after test
Figure 6 The image of the rubber compound

before testing
Figure 9 The image shown a kind of tear
mechanism
Figure 7 The image of pattern formation after the

test Figure 10 The image of a sample after the cryo
fracture test
4. CONCLUSIONS
In this research, mechanical and

tribological properties of conventional
vulcanization rubber compound filled with carbon
175
black HAF330 has been explored. The Universiti Sains Malaysia, Pengenalan
mechanical and tribological results obtained from Penyebatian dan Pemprosesan Getah,
the experiments were comparable with the Penerbit Universiti Sains Malaysia, pp.
commercial tire available in the transportation 14- 16, 23 -28, 66-67 , 89-93, 76-80,
industry. Thus, this is one of the new formulation 113-115, 127, 135– 137, 139, 140 – 141.
that is applicable to manufacture industrial Heinrich G. and Kluppel M. 2007. Rubber
rubber. Furthermore, the evaluation tribological Friction, Tread Deformation, and Tire
test done for both Akron and tribomachine is Traction, pp. 1052-1059
proven quite similar to each other. Therefore, Hofman W. 2000. Rubber Technology Handbook,
tribomachine has been evaluated to be one of the Hanser Publications, London
test for rubber application in the future. The SEM Lindley P.B., Engineering Design with Natural
pictures have shown the wear pattern formation as Rubber, The Malaysian Rubber
previously found by the author (2007) but cryo- Producers’ Research Association, 5th
fracture test has shown that existence of Edition, 1992.
brittleness in rubber surface is non-fibrous. Persson B.N.J. 22 August 2001. Theory of rubber
friction and contact mechanics, Journal
ACKNOWLEDGEMENTS: of Chemical Physics Volume 115,
Number 8, pp. 3840.
This work was supported by the Incentive Grant Stack M.M. 2009. Thesis, Department of
Universiti Sains Malaysia and Fundamental Mechanical Engineering, University of
Research Grant Scheme Strathclyde , Tribology Matter, UK
[203/PMEKANIK/6071192] under Ministry of Power & Process Engineering Issue 1.
Education, Malaysia.
REFERENCES
Bhowmick A.K., Hall M.M. Benarey H.A., and

Decker M. 1994. Rubbers Products
Manufacturing Technology, Inc.
Blow C.M. and Hepburn C. 2000. Rubber
Technology and Manufacture, 2nd
Edition, Butterworth
Braun, Dietrich, 2003.Polymer Synthesis: Theory
and Practices, Fundamentals, Method,
Experiment, 4th Edition, Berlin Springer
Cho J.R., Shin S.W., Yoo W.S. 2004. Crown
Shape Optimization for Enhancing Tire
Wear Performance by ANN, pp. 920-
922.
Dick J.S., 2000, Alpha Technologies, Rubber
Technology: Compounding and Testing
for Performances, Hanser Gardner
Publications, Inc, pp. 1., pp. 2 & 4, pp.
17pp. 126-127.
El-Tayeb N.S.M., Nasir R.M., 4 February
2007Effect of soft carbon black on
tribology of deproteinised and
polyisoprene rubbers, Wear, Volume
262, Issues 3-4, , Pages 350-361
Freakly P.K. and Payre A.R. 1988. Theory and
Practice of Engineering with Rubber,
Applied Science Publisher Ltd.
Gent A. 2000. Engineering with Rubber: How to
design rubber components, Second
Edition, Hanser Publishers
Hanafi Ismail, Azanam Shah Hashim, 1998.
Pusat Pengajian Teknologi Industri,
176
DEVELOPMENT OF ADHESIVE WEAR MODEL USING FEA AND

EXPERIMENTALMETHODS
S.Arasu1, A. Krishnamoorthy1 and K.Srinivasan2

1
Department of Automobile Engineering, Sathyabama University Chennai, India
E-mail: araz_1985@rediffmail.com
1
Department of Mechanical & Production Engineering, Sathyabama University Chennai, India
E-mail: akrish61@gmail.com
2
Department of Mechanical EngineeringAU-FRG institute for CAD/CAM
Anna University, Chennai, India
E-mail: ksrin@annauniv.edu
ABSTRACT
Wear is an important criteria in the field of the wearing surface per unit sliding distance and
Mechanical Engineering. The life and reliability its dimensions are those of length square. For a
of the mechanical components are decided by particular dry or unlubricated sliding situation
the amount of wear that take place in the the wear rate depends on the normal load,
components. Many wear calculations are relative sliding speed, thermal, mechanical, and
validated by experimental methods, but it takes chemical properties of the materials in contact.
long time to complete a calculation and bring There are many physical mechanisms that can
out a meaningful conclusion. This paper contribute to wear and certainly no simple and
„Development of Adhesive wear model using universal model is applicable to all situations.
finite element analysis(FEA) and Experimental
methods‟ focuses in bringing out a wear model
1.1. Mechanism of Adhesive Wear
equation to calculate the adhesive wear volume
in metals like Steel, Aluminum, Copper…etc.
using finite element analysis and experimental The tendency of contacting surfaces is to
adhere to the attractive forces which exist
observation.
between the surface atoms of the materials. If
two surfaces are brought together and then
Keywords:Adhesive wear model, and contact separated, these attractive forces act in such a
analysis, pin-on-disc, Generating rough surface.
way as to attempt to pull material from one
surface on to the other. Whenever a material is
1. INTRODUCTION removed from its original surface in this way, an
adhesive wear fragment is created. The
Wear, the progressive damage involving
mechanism of wear is tangential displacement
material loss which occurs on the surface of a
imposed on one of the bodies. If the force
component as a result of its motion relative to
required to break through the interface of the
the adjacent working parts. It is almost an
materials is larger than the force required to
inevitable companion of friction. Most
attract the materials a break will occur along
tribological pairs are supplied with a lubricant to
this latter surface and a transferred wear particle
avoid the excessive wear and damage which
will be produced. (Ernest Rabiniwicz.,1995)
would be present if the two surfaces were
The adhesive wear can be estimated using a
allowed to rub together. The economic
local coefficient of friction in combination with
consequences of wear are widespread and
a simplified yield model. (Rob Bosman et
pervasive as they involve not only in cost of
al..2011)
replacing parts but also expenses involved in
machine downtime, lost production, and
2. DEVELOPMENT OF MODIFIED WEAR
consequent loss of business opportunities. A MODEL
further significant factor decreases the
efficiency of equipment, which can lead to both Based on the detailed study on various
poor performance and increased energy literatures it is observed that the Archard wear
consumption. equation is the most widely used wear model for
The wear rate of a rolling or sliding contact is the estimation of wear. In order to predict the
conventionally defined as the volume lost from life of complex mechanical systems with
177
hundred thousand operating cycles an the probability that any junction leads to the
equivalent experimental set-up has to be formation of a transferred fragment has been
developed which determines components of postulated to be equal to k. The assumption
wear in the mechanical systems. Considering such that a fragment is hemisphere of diameter
the time and expense for estimating the wear by d, volume V/x of wear per centimeter of
this process, a mathematical model is the sliding is given by the relation
optimum choice for wear simulation based on V k .N . .d 3 kL
Archard‟s wear equation. The first step towards = = (4)
this is to determine the contact pressure
x 12 3p
distribution on the contact surface by using The volume of transferred fragment formed in
finite element procedure. The contact pressure
sliding through a distance x then becomes
obtained by FEA method is compared with the
empirical calculation for a reliable result. This is kLx
used for the determination of wear using V= (5)
Archard‟s wear equation.
3p
This may be considered as the fundamental law
2.1 Archard’s Wear Model of adhesive wear coefficient k. The coefficient
Archard has presented a very plausible of wear (k) is dimensionless. It will be seen that
model of the sliding process which enables us to “(5)” is identical in form with “(1)”. The only
derive quite simply and attach a definite difference being that k/3 has replaced the
meaning to the constant k. In Eqn. 1.1, Archard arbitrary constant c.
assumes that each time two asperities come into
contact to from a junction, there is a constant
probability k that an adhesive fragment will be
formed. Each fragment is assumed to be a
hemisphere of diameter equal to the junction
diameter. If we consider the case of two bodies
sliding under an applied load L and the flow
pressure of the softer of the metals, then the real
area of contact will be given by
L = p. A (1)
If it is assumed that all the junctions are of the

same size, circles of diameter d, and hence of
area ( πd2 / 4) , then the total number „n‟ present
at any instant will be given by
Fig.1 Load- displacement functions of junction.
 .d 2
A = n. Hence,
4 The Fig.1 illustrates (a) as calculated by
Rabiniwicz (1951), (b) as assumed by Archard
L
A= (1953), (c) as measured using an aluminium
p from equation (1) model by Greenwood and Tabor (1955). The
area under each curve representing the total
4L load- distance carried by each junction is about
n = (2)
 . pd 2 the same.
Each junction may be assumed to remain in

existence during a sliding process equal to its 3. DESCRIPTION OF PIN-ON-DISC
diameter after which it is broken and its load- MACHINE (ASTM-G99)
carrying capacity is taken up by a new junction. Probably the most common form of the wear-
Actually the load-displacement function per measuring rig used in tribology laboratories is
centimeter of sliding must be replenished by a pin-on-disc apparatus. It can also be used to
total of 1/d times and total number N of measure the associated degree of wear and
junctions formed per centimeter is given by surface damage. In simplest form the pin is held
n 4L stationary while the disc rotates so that any
N= = (3) particular point on the wear track on the disc
d  . pd 2 comes into contact with the pin. (J.A
Williams.,1994), In a more sophisticated
178
apparatus, the pin may be arranged to move

radially during the course of the experiment so Table:1 Technical Specifications Surface
that it continuously encounters fresh disc Roughness Tester
surface. This arrangement provides single-pass
conditions. However, the speed of the disc must
Particulars Explanation
also be continuously varied if a steady sliding
speed is to be maintained. The actual pin on disc
Skidless method using
machine as shown Fig.2 is used for conducting Measuring
inductive pick-up with
the wear measurement. System
interchangeable stylus
Ra; Rz; Rt; Rp; Rv; Rq;

Measurement
RSm; RmaxD; Material ratio
parameters
curve
Measurable
roughness Ra 0.01 ~ 10 um
range
0.08 mm; 0.25mm; 0.8mm;

Cut off (L)
2.5mm
Fig .2 Pin –on-disc apparatus (ASTM-G99)
Evaluation
4.WEAR EXPERIMENTS AND 1L; 2L; 3L; 4L; 5L
length (LN)
PROCEDURE FOR CONDUCTING WEAR
EXPERIMENTS
A pin-on-disc wear testing machine as per Measuring Horizontal 60mm (vertical
ASTM G99 is used for conducting experiments. Length 0.6mm / 600um)
The surfaces of the equipment are cleaned
preferably with a vacuum cleaner. The disc
Resolution 0.001 um
surface is also cleaned. The pin is mounted in its
holder and is secured firmly in the lever. The
pin height is adjusted so that lever is parallel to Not greater than 6% +/-
disc and tightened well to avoid any slippage Indication error
0.004um
during experiment. The range of wear track
radius is 10 - 75mm. The experiments are
conducted under dry condition. The surface
roughness value is measured after every
experiment using roughness tester. The dry
sliding wear of steel against aluminum with
sliding velocity 0.01 m/s, nominal contact
pressure range 15-40 MPa (Anirban Mahato et
al..2010) values are justified.
4.1Roughness tester Fig.4 En-31 Steel Disc
Roughness tester is used to measure the

roughness value Ra. The roughness tester is
shown in fig.3. The table 1 explains the
technical specification of roughness tester.
Fig. 5 Experiment specimens (pins)
Fig.3 Roughness tester
179
4.2 Disc and pin materials for wear The wear coefficient was calculated using
experiments experimental and finite element method as
shown in table.2
The pins generally are wearing members which
have a lower hardness (L.J. Yang., 1999). The Table.2 wear coefficients (K) from experimental
disc is made by En-31 steel with thickness 8mm and FEA method
and diameter 160mm. The hardness of disc is 65
RC and surface finish‟s rms value is 0.02 as “K"from “K"from FEA
load (kg)
Material
(mm)
WRT
shown in Fig .4. The wear test was conducted experimental method
with steel, aluminium, copper pins of 8mm dia method
and length of 30mm. The pins are shown Fig .5
2 9.01038E-05 9.68128E-05
5. RESULTS OF WEAR CO-EFFICIENT 3 6.63430E-05 6.75027E-05
FROM EXPERIMENTAL AND 53
4 5.83661E-05 5.88496E-05
THEORETICAL
5 6.67797E-05 6.80557E-05
steel
The measured wear value are substituted in 73 8.17251E-05 8.28075E-05
Archard wear equations to find wear co-
53 8.31887E-05 8.81306E-05
efficient of steel, aluminum and copper pins 2
sliding against steel disc. 33 7.63225E-05 8.13514E-05
The Archard Eqn is 13 7.54662E-05 7.66735E-05
2 1.99995E-05 2.12217E-05
(6) 3 2.03993E-05 2.08053E-05
53
4 1.67663E-05 1.68265E-05
Aluminium
Where V is the volume of wear (m3), s is the 5 1.91066E-05 1.93242E-05

sliding distance (m), FN is the normal load, H is
73 1.61917E-05 1.70749E-05
the hardness and K is the wear coefficient. A
wear simulation approach based on Archard‟s 53 1.63559E-05 1.73911E-05
wear law is implemented in an FE post- 2
33 1.51441E-05 1.62323E-05
processor that works in association with a
13 1.46912E-05 1.50553E-05
commercial FE package ABAQUS for solving
the general deformable contact problem 2 3.77228E-05 3.74646E-05
(V.Hegadekatte et al ..2004). In an ABAQUS 3 3.53358E-05 3.64511E-05
contact simulation, surfaces that will be 53
4 3.69149E-05 3.74237E-05
involved in contact must be created on the
copper
various components in the model. Then the pair 5 3.71375E-05 3.74646E-05

of surfaces that may contact each other known 73 4.64980E-05 4.82727E-05
as contact pairs must be identified. 53 3.81537E-05 3.92918E-05
Finally, the constitutive models governing the 2
33 2.72729E-05 2.86440E-05
interactions between various surfaces must be
defined. These surface interaction definitions 13 1.79340E-05 1.85985E-05
include friction (ABAQUS/Standard version).
For rough surfaces, the real contact area is
smaller than smooth surfaces. The ratio Ar/Aa
approaches unity at increased loading (Ulf
Sellgren .,1999). The FEA models are illustrated
fig 6 & fig 7.
Fig.7 Meshed model of disc with pin.
Results are obtained using contact analysis in

Fig.6 Pin FEA model using Ra value. FEM (ABAQUS). Contact pressure distribution
on the pin surfaces of Steel, Aluminium and
180
Copper materials sliding against the steel disc The wear coefficient equation is function of
were determined. The stress plot shows the area load which is given below for steel, copper and
at which contact occurs (real area of contact) for aluminium.
the given load, wear track radius and surface
roughness values. The cumulative value of
y = 0.0001x3 + 0.0072x2 - 0.0461x + 0.1289
contact pressure is used to estimate the normal
load for calculation of wear. The few contact (7)
pressure plot are shown in fig.8, fig. 9 and
y = 0.0009x3 + 0.0073x2 - 0.0181x + 0.0494
fig.10. From the output result file one can find
the real contact areas and contact pressure of (8)
each element. The loads are calculated based on 3 2
summation of real contact areas and summation y = 0.0017x - 0.012x + 0.0247x + 0.0056
of pressure. These loads are substituted into (9)
Archard equation (7) to find the wear
coefficient. The maximum pressure is calculated The results are plotted for wear coefficient
from the equality of total surface pressure and verses load as shown fig.11
working force given in the support
(MiomirJovanović et al..2004). y = 0.0001x3 + 0.0072x2 - 0.0461x +
-3
1X10
0.1289
0.1 R² = 1
0.09
0.08
Wear co-efficient
0.07
y =0.06
-0.0009x3 + 0.0073x2 - 0.0181x Steel pin
0.05 + 0.0494
0.04 R² = 1 Aluminium
0.03 Copper
0.02
0.01 Poly. (Steel
pin)
0
19.62 29.43 39.24 49.053 2
y = 0.0017x - 0.012x + 0.0247x +
Load(N) 0.0056
R² = 1
Fig.11 load Vs wear co-efficient (Steel,

Aluminium, Copper)
Fig .8 Contact pressure plot steel
The wear coefficient equation is function of
velocity which is given below for steel, copper
and aluminium.
y = 2E-06x3 - 2E-05x2 + 4E-05x + 6E-05

(10)
y = 0.0007x3 - 0.0053x2 + 0.0029x + 0.0483
(11)
3 2
y = 0.0011x - 0.0104x + 0.0235x + 0.0019
(12)
Fig .9 Contact pressure plot Aluminium
The results are plotted for wear coefficient
verses velocity as shown fig.12
Fig .10 Contact pressure plot copper
181
L.J. Yang, “Pin-on-disc wear testing of tungsten

1X10
-3
y = 2E-06x3 - 2E-05x2 + carbide with a new moving pin
0.09 4E-05x + 6E-05 technique”, Wear 225–229
0.08 R2 = 1
0.07 Steel pin
_1999. 557–562
Wear co-efficient
y= 0.0007x3 - 0.0053x2 + 0.0029x +

0.06
0.05 0.0483 Aluminium M.KathrynThompson,“Methods
R² = 1
0.04 Copper for Generating Rough
0.03 Poly. (Steel
pin)
Surfaces in ANSYS” (2006)
0.02
0.01 MiomirJovanović PredragMilić,
0
y = 0.0011x 3 - 0.0104x2 + 0.0235x + DankoMijajlović “Approximate
3822.271062
2775.073511
1727.875959
680.6784083
Sliding Velocity (mm/sec)
0.0019 contact models of the rolling
R² = 1
supports”, FactaUniversitatis
Fig.12 velocity Vs wear co-efficient (Steel, (Series: Mechanical Engineering
Aluminium, Copper) Vol. 2, No 1, 2004, pp. 69 – 82]
V.Hegadekatte, N. Huber and O. Kraft (2004)
7. CONCLUSION “Finite element based
simulation of dry sliding wear
The wear experiments were carried out for three Ulf Sellgren, Stefan Björklund and
different materials which are steel, copper and SörenAndersson“A Contact
aluminium against steel. Model for Rough Surfaces”
Adhesive wear is measured for all the three (NAFEMS World Congress,
combinations. Rhode Island, USA, April 25-
The wear model is developed using curve fitting 28, 1999)
methods with respect to different loads and
velocities. NOMENCLATURE
FEA model was developed using contact
analysis to measure the wear. w Wear dimension
The wears were obtained through empirical
relation and the FEA model was compared and P Applied load (N)
found that it had close agreement within 5%.
 Coefficient of friction
t Time(minutes)
REFERENCES
V Wear volume (mm3)
ABAQUS,Standard version (6.5-1)
Users Manual contact S Sliding distance (mm)
algorithm”
W Normal load (N)
Ernest Rabiniwicz, “Friction and K wear coefficient
wear of materials”(1995) H Hardness of the material (N/mm2)
J.A Williams, “Engineering c Non dimensional constant
Tribology” (1994)
Katherine James C. Wyant, WRT Wear track radius(mm)
“ measurement of surface roughness”, 10 ASTM American Standard Testing
September 1990 / Vol. 29, No.
26 / Applied Optics. Aa Apparent Area (mm)
Ar Real contact Area (mm)
Rob Bosman , Dirk Jan Schipper“Transition
from mild to severe wear
including running in
effects”(2011)
Anirban Mahato, Thomas A. Perry,

Vikram Jayaram, S.K. Biswas
“Pressure and thermally induced
stages of wear in dry sliding of
a steel ball against aluminium–
silicon alloy flat”(2010)
182
Effect of T6 Heat Treatment on Wear Properties of AlN Reinforced Al-Si alloy

Composites
M.N Wahab1, M.J. Ghazali2, and A.R. Daud3

1
Fabrication and Joining Section, Universiti Kuala Lumpur-MFI,
43650, Bangi, Selangor, Malaysia
E-mail: mahamadnoor@mfi.unikl.edu.my
2
Department of Mechanical & Material Engineering,
Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia
E-mail: mariyam@eng.ukm.my
3
School of Applied Physics, Faculty of Science and Technology,
Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia
E-mail: ard@ukm.my
ABSTRACT most promising structural materials due to their

high specific strength and stiffness, which are
Dry wear properties of Aluminum nitride (AlN) widely used in automobile industries for
reinforced aluminum silicon alloy (Al-Si) structural components in engines and other parts
studied by using a pin-on-disc configuration (Rittner, 2000). Various manufacturing
tester. In this work different weight percentage processes are available in the industry for
(wt %) of 0, 2, 5, 7 and 10 of AlN was added in discontinuous metal matrix composites.
the metal matrix composites. The composites Amongst them, stir casting method has drawn
were prepared by a stir casting process which considerable research interest due to its
was then aged with a T6 condition prior to the simplicity, flexibility and applicability to large
wear test. Detailed analyses on the quantity production (Hasyim et al, 1999). In
morphologies were conducted using a scanning addition, in principle it allows a conventional
electron microscope (SEM) with the aid of an metal processing route to be used, hence
energy dispersive x-ray analyser (EDX). It was minimises the final cost of the product. Particles
clearly observed that the hardness improved reinforced aluminum alloy composites have
from 70 to 240% for the aged samples shown significant improvement in the
compared to un-aged samples. The specific terminological properties including sliding and
wear rate or the K’ valued was 2.06 x 10 -4 abrasive wear resistance and seizures
mm3N-1m-1 and 1.35 x 10-4 mm3N-1m-1 for Al-Si resistances. Better wear resistance of particulate
alloy and Al-Si/10%AlN respectively and reinforced MMCs is normally due to the hard
improved to 70% and 60% respectively after the ceramic particle content which finally gives a
heat treatment. A surface investigation by the protection from further surface damage (Geng et
scanning electron microscope (SEM) revealed al., 2006). The aging behaviour of
that thick mixed mechanical layers (MML) with discontinuously reinforced metal matrix
wear delamination occurred on the Al-Si alloys composites has been a subject of great interest,
while thin layers (MML) were observed in the which is beneficial to optimise the aging
AlN reinforced composites. treatment and provide the experimental and
theoretical information for designing the
Keywords: Al-Si alloy, AlN, mixed mechanical properties of the composites (Sheu & Lin,
layer, reinforced particles 1997). Many works on SiCp/Al alloys and
Al2O3/Al alloy composites have been reported
1. INTRODUCTION (Sevik & Kurnaz 2006; Daoud & Reif 2002) but
very few reports on the use of AlN as
The A good combination of high strength and reinforcement for Al alloy composites (Vicens
ductility of the aluminum based metal matrix et al., 2002). Limited studies was done so far to
composites (MMCs) has introduced the material investigated the behaviour of AlN reinforced
to a wide area of possible advanced matrix alloy and mostly concentrated in the
applications. Aluminum alloys are amongst the electrical field (Sheppard, 1990), thin films and
183
semiconductor application (Shimada & Hayashi,

2005). With comparison to other ceramic
materials, AlN provide the lowest densities,
high specific modulus, and lowest thermal
expansion. Its thermal expansion of 4.6 x 10-6
m/ºC was close to Si and may offer a potential
use for both monolithic and composites
materials (Karamuto & Takada, 2003). Thus the
objective of this research is to investigate the
effects of AlN particles in Al-Si alloy
particularly on the wear behaviors before and
after the heat treatment.
Figure 1 the actual size of reinforced AlN
particles
2. METHODOLOGY
Some sample of the matrix Al-Si alloy were
A commercially available Al-Si alloy series subjected to T-6 type heat treatment which
(AC3A.2) in Table 1 with chemical involved solutionising at 540 ˚C followed by
composition, 88 wt.% Al, 11.1 wt% Si, 0.42 quenching and artificial ageing at 180 ˚C. The
wt.% Fe, 0.02 wt.% Cu, 0.02 wt.% Zn and other soaking time was fixed at 4 hours. The macro
elements as impurities was used as the matrix hardness tests were taken by using a DXT
for the composite. Aluminum nitride (AlN) Shimadzu hardness tester with a 1/16 inch steel
powders from Aldrich manufacturer with a ball indenter. Each sample was taken 10
purity of >98% and the size of <10 µm were indentation points with 100 kg indenting load
used as the reinforcement materials. Figure 1 and dwelling time of 15 second. Prior to the
shows the actual AlN size which observed by wear test, as-cast Al-Si alloy sectioned and
the scanning electron micrograph (SEM). Al-Si metallographically polished and etch with
alloys with and without the reinforcement were Keller's reagent (Rao et al., 2010). Testing
prepared by a stir casting process. A small parameter was fixed i.e. disk speed 400 rpm,
amount of magnesium i.e. 2.5 wt% was melted sliding diameter of 48 mm and a load range of
together with the Al-Si alloy. The purpose of 25N to 70. The sliding distance was 1-5 km
adding the magnesium addition was to improve with a fixed velocity of 1 ms-1. The wear test
the wetting properties of the melt and to ease was carried out against steel made from a
binding process between the molten metal and hardened carbon steel grade EN-31. Before and
the reinforced particles (Hashim et al., 2001). after each test, sample was weighed to
During casting process, melting temperature determine the volume loss. Worn surface for
was fixed at 750 ºC, held for 30 minutes before each sample were cleaned with acetone prior to
aluminum alloy melted completely. Aluminum SEM and EDX analyses.
dross was then removed from the surface of the
molten metal. Small amounts of particulate
aluminum nitride was preheated to 750 ºC and 3. RESULT AND DISCUSSION
added continuously to the molten metal through
the side of a vortex created by the stir impeller. The microstructures of as-cast and aged Al-Si
The Al-Si was then casted via a bottom furnace. alloys, and composites reinforced with 10 wt%
AlN, aged at 180oC for 4 hrs are shown in
Table 1: Chemical composition of Al-Si alloy Figure 2. AlN particles seemed to be randomly
trapped after the casting and were observed at
the boundaries of silicon grains shown in Figure
Elements wt%
2(b). The formation of AlN particles entrapment
Si 11.10 can be explained by the Al dendrite growth
Fe 0.420 during the solidification process whereby the
Zn 0.020 AlN was pushed to the outer side of the last
Cu 0.020 freezing point of eutectic liquid, thus the AlN
Sn 0.016 particles are seen surrounded by the Si. This
was due to AlN particles and Si element in the
Mg 0.011
matrix that had the same thermal expansion
Al Balance properties (Baik & Drew, 1996). An effect on
aged treatment had transformed the Si grains
with equiaxed shapes while the Al dendrites
184
tend to be more globular and bigger in sizes. agglomerated as discussed by previous

The precipitated Si in the Al matrix is shown in researchers (Hashim et al., 1999).
Fig. 2(c) and was reported in finding previously
(Curle & Ivanche, 2010). The AlN particles
were seen similar with the as-cast alloys
whereby the trapped AlN closed to the
precipitates Si (Fig. 2(d)). The presence of the
AlN in the Al-Si matrix was supported by an
EDX as shown in Figure 2(e). Similar behaviour
was reported by Longtao et al., ( 2005) in which
they observed a formation of grain dislocations
and precipitates in aged 2024Al of sub-micron
Al2O3/2024Al composites.
Figure 3 Hardness properties of as-cast and aged

Al-Si/AlN composites
Specific wear rates (K’) for as-cast and

aged Al-Si/AlN composites were studied against
the applied loads as shown in Figure 4. Figure
4(a) shows the wear rate of Al-Si alloy, Al-
Si/7%AlN and Al-Si/10%AlN at 25N. The
higher wear resistance of AlN reinforced
composites was due the dominant effect of the
hardness (Karamuto & Takada, 2003) of the
Figure 2 Morphologies and EDX spectrum for AlN particles enhancing the wear resistance of
Al-Si/AlN composites with (a) as cast the composites. At 50N, the wear rates were still
Al-Si alloy, (b) as-cast Al-Si/10%AlN, similar to the lower loads. Beyond 50N the wear
(c) and (d) for heat treated Al-Si alloy rates changed drastically and increased with the
and Al-Si/10%AlN respectively. load complying the Archard equation of wear
(Suresh et al., 1993), which stressed that the
Figure 3 shows the hardness values for wear rates were low at the lower loads and
Al-Si alloys and their composites for aging and increased with load at a constant ratio. An
heat treated AlN composites. Significant observation of the heat treated sample showed a
increase in the hardness of the Al-Si matrix reversed behaviour of wear resistance by
alloy can be seen by the addition of AlN showing stagnancies of wear rates for both load
reinforced particles. An increased in hardness (Fig. 4(b)). The value of wear rates or K’ at
value indicated that the particulates presence in 70N reduced tremendously to 5.97 x 10 -5 mm3N-
1 -1
the matrix had improved the overall hardness of m and 5.20 x 10-5 mm3N-1m-1 for as-cast Al-Si
the alloys. For as-cast Al-Si alloy and alloy and Al-Si/10%AlN respectively. As for K’
composites, hardness increased from 17 HRB at the same load before aging was at 2.06 x 10 -4
for Al-Si matrix alloy to 44 HRB and 40 HRB mm3N-1m-1 and 1.35 x 10-4 mm3N-1m-1 for as-
for Al-Si/7%AlN and Al-Si/10%AlN cast Al-Si alloy and Al-Si/10%AlN
respectively. An increase trend was also respectively. It was noted that the precipitation
recorded for the heat treated sample with 59 hardening of Si needles and the re-arrangement
HRB for Al-Si alloy to 78 HRB and 75 HRB for of the AlN particles due to aged treatment as
treated Al-Si/7%AlN and Al-Si/10%AlN show in Fig. 3 improved the wear properties of
respectively. Precipitation hardened of Si grain the Al-Si/AlN.
may attributed to the increased hardness in aged
Al-Si matrix alloy as previously reported by
Harun et al., (1996). As for the aged AlN The cross section of the worn surface in Figure
reinforced Al-Si matrix alloy, the properties of 5 exhibits various types of mixed mechanical
AlN itself may contributed to the increased in layers (MML). A close scrutinised of as-cast Al-
hardness. A small decreased in hardness for as- Si alloy shown piled large loosed debris. Wear
cast and aged Al-Si/10%AlN was due to the mechanisms of delamination was suspected in
presence of micro pores which associated with the as-cast Al-Si alloy (Fig. 5(a)). The
delamination wear was initiated with a
185
formation of abrasion caused by the hardened

counter-face onto the pin surface. As a result,
the soft was scarified with formations of craters
at the delamination grooves. The result was
supported with higher wear rates as shown by
the curve formation of Al-Si alloy in Fig. 4(a).
Similar behaviour of wear regime was also
reported by Lim et al., (2003), Kok and Ozdin
(2007) for wear test at low speed of 1 ms-1 to 3
ms-1 and load ranges from 10-100N. As for the
un-treated Al-Si/10%AlN composites (Fig.
5(b)), shallow and thin mixed mechanical layers Figure 4 Wear rates K’ against various load for
(MML) were observed on the worn surface. The Al-Si/AlN composites with (a) as-cast
composites and (b) heat treated
presence of hard AlN particles seen in Figure
2(b) reduced the extent of penetration of the composites
sample surface, therefore protecting the softer
Al alloy. This similar result was reported by
Sahin & Acular (2003) and Ahlatcia et al.,
(2006) in their observation of the wear
behaviour of SiC and aluminum oxide
reinforced Al matrix composites.
SEM morphologies for heat treated Al-Si

alloy showed similar structures as the as-cast
alloy with large and deep MMLs (Fig. 5(c)).
The Al-Si surface was found with large craters
and wide delamination grooves. The
precipitation Si grains and others inter-metallic
compounds improved the hardness, resulting
significant increased in wear resistance. In this
case, it enhanced wear resistance in both Al-Si
matrix and composites. Heat treated Al-Si alloy
improved the ductility of Al-Si material which Figure 5 Cross section of as-cast and heat
was indicated by then sign of material extrusion treated worn pin surface with (a) as-cast
at the end of pin surface (Fig. 5(c)) thus Al-Si alloy; (b) as-cast Al-Si/10%AlN
reducing a tendency to wear. As for then heat composite; (c) heat treated Al-Si alloy and
treated Al-Si/10%AlN composite in Figure 5(d) (d) heat treated Al-Si/10%AlN composite
the cross section showed precipitated Si and
reinforced AlN particles dominated the Al-Si
microstructure with thin MML layer on top. The 4. CONCLUSION
smooth contour may be explained by the
presence of hard and brittle AlN particles in the A stir casting process was successfully
Al-Si alloy. When cracks occurred the AlN fabricated Al-Si composites reinforced with
particles were ejected from the surface which AlN particles. AlN particles seemed to be
tends to be a buffer zone between the pin and randomly trapped after the casting and were
the counter-face, thus reducing the wear rates. observed at the boundaries of silicon grains. An
This argument was supported by the wear aged treatment had transformed the Si needles
resistance of the treated Al-Si/10%AlN in to equiaxed shapes due to precipitation
Figure 4(b) with more than two fold increase as hardening. Hardness properties improved 70%
compared to the as-cast composites. Similar to 240% for the aged composites compared to as
behaviour was observed in Al2O3 reinforced cast composites. Specific wear rates or K’ at
particles where the debris acted as a buffer of 70N were 2.06 x 10-4 mm3N-1m-1 and 1.35 x 10-4
the matrix wear resistance (Kok & Ozdin 2007). mm3N-1m-1 for as-cast Al-Si alloy and Al-
Si/10%AlN respectively and it reduced 70% and
60% after the heat treatment. SEM morphology
showed thick layer of MML for as-cast Al-Si
alloy due to delamination while for aged
samples the combination of delamination wear
and extrusion dominate its worn surface. A heat
186
treated Al-Si/AlN composites observed with Key engineering materials, 247:467-

presence of hard AlN particles mixed with wear 472.
debris act as buffer zone thus protect the Al-Si Lim, C.Y.H., Lim, S.C. & Gupta, M. 2003.
surface. Wear behaviour of SiCp-reinforced
magnesium matrix composites. Wear,
1-6:629-637.
REFERENCES Longtao, J., Min, Z., Gaohui, W. & Qiang, Z.
2005. Aging behaviour of sub-micron
Ahlatcia, H., Kocera, T., Candan, E., Al2O3p/2024Al composites. Materials
Cimenoglu, H. 2006. Wear behaviour Science and Eng. A, 392:366-372.
of Al/(Al2O3p + SiCp ) hybrid Rao, R.N., Das, S., Mondal, B.P. and Dixit, G.
composites. Tribology International, 2010. Effect of heat treatment on the
39:213-220. sliding wear behaviour of aluminium
Baik, Y. & Drew, R.A.L. 1996. Aluminum alloy (Al–Zn–Mg) hard particle
nitride: Processing and application. composite. Tribology International
Key engineering materials, 122- 43:330-339.
124:553-570. Sevik, H. & Kurnaz, H.C. 2006. Properties of
Sheu, C.Y. and Lin, S.J., 1997. Aging behavior alumina particulate reinforced
of Sic,-reinforced AA7075, aluminum alloy produced by pressure
Composites, Journal of Materials die casting. Materials & Design 27:
Science, 32: 1741-1747. 676-683, 2006.
Curle, U.A. and Ivanchev, L. 2010. Wear of Sahin, Y. and Acular, M. 2003. Production and
semi-solid rheocast SiCp/Al metal properties of SiCp-reinforced
matrix composites. Trans. aluminum alloy composites.
Nonferrous Met. Soc. China 20:852- Composites Part A, 34(8): 709-718.
856 Sheppard, L.M. 1990. Aluminum nitride: a
Daoud, A. & Reif, W. 2002. Influence of Al2O3 versatile but challenging material.
particulate on the aging response of American ceramic society bulletin,
A356 Al-based composites. Journal 69:108-112.
of Materials Processing Technology Shimada, S. & Hayashi, S. 2005. Preparation of
123 (2):313-318. monolithic AlN and composite TiN–
Geng, L., Zhang, X.N., Wang, G.S., Zheng, Z.Z. AlN powders and films from
and Zu, B. Effect of aging treatment precursors synthesized by electrolysis.
on mechanical properties of Materials Science and Engineering B,
(SiCw+SiCp)/2024Al hybrid nano- 122:34-40.
composites. Trans Nonferrous Met. Suresh, S., Mortensen, A. & Needleman, A.
Soc. China, 16(2): 387-391. 1993. Fundamentals of Metal-Matrix
Harun, Mohd, Talib, I. A. & Daud, A. R. 1996. Composites. Stoneham, USA:
Effect of element additions on wear Butterworth-Heinemann.
property of eutectic aluminium-silicon Vicens, J., Chédru, M. & Chermant, J. L. 2002.
alloys. Wear 194 (1-2):54-59. New Al-AlN composites fabricated by
Hashim, J., Looney, L. & Hashmi, M. S. J. squeeze casting: interfacial
2001. The wettability of SiC particles phenomena. Composites Part A:
by molten aluminium alloy. Journal of Applied Science and Manufacturing,33
Materials Processing Technology 119 (10):1421-1423.
(1-3):324-328.
Hashim, J., Looney, L. & Hashmi, M.S.J. .
1999. Metal matrix composites:
production by the stir casting method.
Journal of Materials Processing
Technology, 92-93:1-7.
Kok, M. & Ozdin, K. 2007. Wear resistance of
aluminum alloy and its composites
reinforced by Al2O3 particles. Journal
of Materials Processing Technology,
183:301-309.
Kuramoto, N. & Takada, K. 2003. Application
of aluminum nitride (AlN) ceramics.
187
Paper Reference ID: RTC053
EROSION-CORROSION BEHAVIOR OF 316-SS IN SEAWATER SIMULATED

ENVIRONMENT AT VARIOUS IMPINGEMENT ANGLES
H.M. Ghasemi1,*, M. Karimi1, A. Pasha1,2 and M. Abedini1

1
School of Metallurgy and Materials Engineering, University of Tehran, Tehran, Iran.
*
Corresponding author: hghasemi@ut.ac.ir
2
Corrosion Department of Iranian Research Institute of Petroleum Industry, Tehran, Iran.
ABSTRACT 1999). Therefore, mechanisms of interaction

must be understood and quantified.
The effect of sand erosion on corrosion behavior The measurement of the synergistic
of 316-stainless steel in a simulated seawater effect of erosion and corrosion is carried out in
environment was studied in a jet slurry accordance with ASTM-G-119. This standard
apparatus. 10 wt.% SiO2 particles with average provides a means for computing the increased
size of 250-500 μm were used as the erodent in wear loss rate attributed to synergism or
a 3.5 wt.% NaCl solution. The tests were carried interaction that may occur in a system when
out in accordance with ASTM G 119 at three both wear and corrosion processes coexist. The
slurry impinging angles of 25°, 55°, and 90° at main idea of this standard based on division of
an impinging velocity of 6 m/s. Pure erosion, total metal loss T, into three subparts: W0, C0
pure corrosion and erosion–corrosion tests were and S by following equation:
performed and the synergistic effect was
calculated. Positive synergistic effect was T = W0 + C0 + S (1)
observed at all examined impinging angles. The
results showed that the highest pure erosion and In this equation, W0 indicates mechanical
erosion-corrosion rate was occurred at an erosion wear rate (i.e., with cathodic protection
impingement angle of 25°. The results also in corrosive media), C0 designates the
showed that corrosion increased the erosion rate electrochemical corrosion rate (i.e., no wear),
by about 57, 37 and 57 percent at the and S represents the synergistic effect between
impingement angles of 25, 55 and 90 degree, the two processes.
respectively. Scanning electron micrographs of Identifying the erosion-corrosion
the eroded surfaces showed higher plastic mechanism is highly dependent on the erosion-
deformation and cutting at the impinging angle corrosion parameters such as erosive media
of 25°. (Wang et al., 2000, Ping et al., 2006, Stack et
al., 2004), impingement angle and velocity
Keywords: Erosion-corrosion, 316 stainless (Lopez et al., 2005, Burstein et al., 2000, Sasaki
steel, Impinging angle, Synergistic effect et al., 1996, Stack et al., 1999, Al-Bukhaiti et
al., 2007, Jana et al., 2005, Wharton et al.,
1. INTRODUCTION 1995), cathodic protection potential (Yao et al.,
1995), temperature (Neville et al., 1996, Hu et
Erosion-corrosion is a serious problem for al., 2005) and solid loading (Meng et al., 2007).
material durability in many industrial processes Passivating materials such as 316-SS, are
such as chemical, mining, hydraulic and protected against corrosion by a dense and
metallurgical industries that handles slurry passive oxide film of a few nanometres on their
components such as pumps, agitators, valves surfaces (Lo et al., 2009). In the erosion-
and pipelines (Wood, 2006, Toro et al., 2001, corrosion processes, mechanical loading and
Neville et al., 2002). Two metal degradation impinging of solid particles as well as the
mechanisms; namely erosion and corrosion electrochemical mechanisms can destroy this
work together and their synergistic effect cause passive layer. It seems that corrosion rate
severe metal loss in such cases. As a result of increases with removal of passive layer by
this synergistic effect between these erodent particles and corrosion promotes
mechanisms the overall metal loss of a erosion by dissolving the work-hardened layer
component can be greater than the sum of the on the specimen surface as well as roughening
metal loss due to each mechanism acting the surface (Yao et al., 1995, Neville et al.,1996,
separately (Lopez et al., 2007, Bermudez et al., Hu et al., 2005, Meng et al., 2007).
2005, Berget et al., 2007, Hodgkiess et al.,
188
Surface roughness is an effective as the impingement angle decreased. Therefore,

parameter that may affect the pitting potential of the erosion rate was normalized based on the
304L stainless steel. An increase in the number of impacted particles on the sample
smoothness of the metal surface causes increase surface. In order to obtain the normalized
in the pitting potential (Sasaki et al., 1996). values, the erosion rate was divided by the sin θ,
Erosion causes the change in the surface where θ is the impingement angle,
roughness and in the low angle impingements;
the surface roughness of 304L stainless steel is
higher than the high angle impingement. This
could increase the erosion-corrosion rate of this
material at low angle impingement. In the other
hand, the reason for an increase in the total
materials loss of 304 L stainless steel during
erosion-corrosion may be the breakdown the
passive layer of metal and formation a rough
surface on the metal (Sasaki et al., 1996).
In this study, we aim to evaluate the
erosion-corrosion behavior of 316-SS in
different impinging angles. This material
Figure 1 Schematic of erosion-corrosion jet
compromises the material cost with good
impingement rig, 1: nozzle, 2: sample, 3:
corrosion and erosion behavior due to 2 wt.%
reference electrode in capillary, 4: counter
molybdenum in chemical composition of the
electrode, 5: pump, 6: flow control valve, and
alloy. Synergistic effects of erosion and
7: potentiostate.
corrosion were evaluated using ASTM G-119
standard test method. Tests were carried out at
For each impingement angle, two distinct
different impingement angles in a 3.5 wt. %
tests were performed in a similar solution to
NaCl solution containing 10 wt.% SiO2.
obtain the weight loss of material due to the
pure erosion (W0) and the erosion-corrosion (T).
2. EXPERIMENTAL PROCEDURE
The pure erosion test (i.e., to obtain W0) was
carried out using an applied cathodic protection
AISI 316-stainless steel disks with a diameter of
voltage of -1300 mV. In these tests, the counter
5 mm (i.e., an exposed area of 0.196 cm2) were
electrode was 316-SS and the reference
used as the test samples. The specimens were
electrode was a saturated calomel electrode
ground using 1200 SiC paper. Erosion and
(SCE) in a capillary. The polarization curve of
erosion-corrosion tests were performed using a
the 316-SS in the static 3.5% NaCl solution was
slurry erosion impingement rig that could pump
also obtained to measure the static corrosion
the erosive-corrosive media through a nozzle on
rate (C0). The synergistic effect (S) was then
the sample surface at controlled velocities and
calculated using Eq. (1). Finally, the eroded
impingement angles at ambient temperature.
surfaces were characterized using a scanning
Schematic of this machine is shown in Figure 1.
electron microscope (SEM).
The nozzle with a diameter of 8 mm (i.e., an
area of 0.50 cm2) could erode the whole area of
the sample (i.e., a constant area of 0.196 cm2) at
any impingement angle. The distance between
The polarization curve of the 316-SS in static
the jet nozzle and the specimen surface was
3.5% NaCl solution (i.e, no mechanical erosion
adjusted to be about 5 mm. A slurry electrolyte
of the surface) is shown in Figure 2. The
solution of 3.5 % sodium chloride (NaCl)
corrosion rate (C0) of the 316-SS in NaCl
containing 10 wt. % SiO2 with average size of
solution in the static condition was calculated
250-500 m as eroding particles was used in the
from the polarization curve (Berget et al., 2007)
tests. Impinging angle could be adjusted by
and showed a value of about 0.006 mg/cm2.h.
rotation of specimen in front of the impinging
The anodic branch of Figure 2 indicates that the
nozzle. The tests were performed at
alloy behaved as a passive material in the static
impingement velocity of 6 m/s and
condition.
impingement angles of 25°, 55°, and 90° for 45
Figure 3 shows the mechanical only
minutes. After each test, samples were dried
erosion rate (W0) of 316-SS in 10 wt. % SiO2
and weighted using a precision balance with an
NaCl solution at various impingement angles
accuracy of 0.1 mg. At a specific concentration
with application of cathodic protection
of impinging particles, the number of impacted
potential. This figure shows that the impinging
particles on the surface of the sample decreased
angle plays a major role in the erosion rate of
189
316-SS in the NaCl solution. The highest 2000, Sasaki et al., 1996). At 90o impinging
erosion rate was obtained at impingement angle angle, the impact of the particles could
of 25o and decreased as the angle increased. The introduce higher work hardening and lower
more cutting mode behavior at low impinging erosion of the eroded surface as shown in Figure
angles compared with the more impact mode 3.
behavior at higher impinging angles resulted in
higher erosion of this ductile alloy at low angles (a)
(Hutchings, 1992).
1500
1000
500
Potential (mV), SCE
0
Cutting edges
-500
-1000
(b)
-1500
-3 -2 -1 0 1 2 3
Log i (mA/cm2)
Figure 2 Polarization curves for 316-SS in a
NaCl solution in static condition.
(c)
Figure 3 Normalized pure erosion rate (W0) per

unit mass of impingement particles of 316-SS in
a 3.5 wt. % NaCl solution containing
10 wt.% SiO2 with application of cathodic
protection voltage.
Figure 4 shows the SEM micrographs of

the eroded surfaces of 316-SS alloy in 10 wt. %
SiO2 NaCl solution at different impinging Figure 4 SEM micrographs of eroded surface
angles. The cutting edges due to the impact of of 316-SS in 10 wt. % SiO2 NaCl solution at
erosive particle on the surface of samples could impingement velocity of 6 m/s after 45 minutes
be observed. The pitting-like dark locations in under impinging angles of, a) 25o, b) 55o
the micrographs could be probably due to the and c) 90o.
contrast made by emission of lower electrons
from the locations from the behind of the edges Figure 5 shows the erosion-corrosion
of the deformed surfaces. These locations were rate (T) for 316-SS in 10 wt.% SiO2 NaCl
also attributed to the local corrosion (pitting) of solution. The figure shows that the erosion-
the eroded surface due to the breakup of the corrosion rate decreased as the impingement
protective oxide layer (Rajahram et al., 2009). angle increased, the same trend as was observed
Figure 4 shows more cutting mode of erosion at for pure erosion shown in Figure 3. However,
lower angles of 25o and 55o than 90o in the ratios of removal rate of the material at
agreement with other works (Burstein et al., different angles were very different during the
190
erosion-corrosion and the pure erosion tests. For corrosive media. These results are in agreement
example, this ratio at impingement angle of 25o with other works (Burstein et al., 2000).
to 55o was about 2.16 and 1.47 for the erosion- Figure 6 also shows that at impingement
corrosion and the pure erosion tests, angles of 25o, 55o and 90o about 57%, 37% and
respectively. This behavior could be explained 57% of total material loss (T) was related to the
by the interaction between erosion and synergistic effect (i.e., S), respectively. At
corrosion, i.e., the synergistic effect. higher impingement angle of 55o, there was less
cutting of the surface (Hutchings, 1992) leading
to a less effective area for corrosion and,
therefore, lower synergy than impingement
angle of 25o. At the impingement angle of 90o,
there was less cutting and more work hardening
due to the direct impact of the particles on the
eroded surface leading to a lower erosion of the
material. The formation of the passive film on
the stainless steel due to corrosion and direct
breakage of the brittle film at the high
impingement angle of 90o helped to increase the
synergistic effect compared to the test at the
Figure 5 Normalized erosion-corrosion rate (T)
impingement angle of 55o.
of 316-SS in a 3.5 wt. % NaCl solution
containing 10 wt.% SiO2.
The erosion rate due to the synergistic

effect between erosion and corrosion were
calculated from Eq. (1). The results are
presented in Figure 6, as a function of
impingement angle. The erosion-corrosion (T)
and the sum of pure erosion (i.e., no corrosion)
and corrosion (i.e., no erosion) are also shown
in the figure as W0+C0. It should be mentioned
that the value of corrosion rate, C0, was constant
and had a very low value compared with W0
(Figure 3). Figure 6 shows that the total
erosion-corrosion rate (T) was greater than the
sum of pure erosion and corrosion (W 0+C0), Figure 6 Synergistic effect (S) at various
indicating positive synergy effect at all studied impingement angles in erosion-corrosion tests
impingement angles. This could be due to the of 316-SS in a 3.5 wt. % NaCl solution
effect of erosion on corrosion as well as effect containing 10 wt.% SiO2 calculated from Eq.
of corrosion on erosion. High synergistic effect (1). The total rate during erosion-corrosion (T)
reveals that although the static corrosion rate and the sum of pure erosion and static corrosion
(C0) of the alloy is very low, however, it could (W0+C0) are also presented.
increase the total weight loss in erosion-
corrosion condition by affecting the erosion 4. CONCLUSION
mechanisms and, therefore, is an important
parameter in the erosion-corrosion tests. 1- The erosion-corrosion rate of 316-SS in
The synergy effect could be due to the 3.5% NaCl solution with 10 wt. % SiO2
continuous corrosion of the eroded surface and particles was highly dependent on the
vice versa. Figure 6 shows that the highest impingement angle of eroding particles.
synergistic effect (S) occurred at impingement The lower the impingement angle, the
angle of 25o. Figure 4 shows more cutting edges higher the rate.
on the eroded surface at this angle. The 2- In general, the synergistic effect was very
corroding solution could be stagnant just under important parameter in erosion-corrosion of
these edges and might cause some local 316-SS in 10 wt.% SiO2 within the NaCl
corrosion and pitting or loosening of metallic solution. Positive synergistic effect was
bonds. The detachment of these edges due to observed at all impingement angles. The
the subsequent impact of the SiO2 particles and highest synergic effect was observed at
erosion of the pitted areas could further increase impingement angle of 25o. More than half
the synergy of the surface damage in the of total weight loss in erosion-corrosion
condition was contributed to the interaction
191
between erosion and corrosion, i.e., of austenitic and martensitic high

synergistic effect. nitrogen stainless steels. Wear, 263:
3- Morphology of the mechanically eroded 347–354.
surface, corrosion of the eroded surface and Lo, K.H., Shek, C.H. and Lai, J.K.L. 2009.
subsequent erosion of the corroded surface Recent developments in stainless
determined the extent of synergy effect of steels. Journal of Materials Science and
erosion and corrosion on the surface Engineering R, 65: 39–104.
damage at various impingement angles. Meng, H., Hu, X. and Neville, A. 2007. A
systematic erosion–corrosion study of
two stainless steels in marine
REFERENCES conditions via experimental design.
Wear, 263: 355–362.
Al-Bukhaiti, M.A., Ahmedb, S.M., Badran, Neville, A., Reyes, M. and Xu, H. 2002.
F.M.F. and Emarab, K.M. 2007. Effect Examining corrosion effects and
of impingement angle on slurry erosion corrosion/erosion interactions on
behaviour and mechanisms of 1017 metallic materials in aqueous slurries.
steel and high-chromium white cast Tribology International, 35: 643–650.
iron. Wear, 262: 1187–1198. Neville, A. and Hodgkiess, T. 1996. An
Berget, J., Rogne, T. and Bardal, E. 2007. assessment of the corrosion behavior of
Erosion-corrosion properties of high grade alloys in sea water at
different WC-Co-Cr coatings deposited elevated temperature and under a high
by the HVOF process-influence of velocity impinging flow. Journal of
metallic matrix composition and spray Corrosion Science, 38(6): 927-956.
powder size distribution. Surface & Ping, L., Qi-zhou, C., Bo-kang, W., and Xian-
Coating Technology, 201: 7619-7625 zhong, Z. 2006. Effect of aging
Bermudez, M.D. and Carrion, F.J. 2005. temperature on erosion-corrosion
Erosion-corrosion of stainless steels, behavior of 17-4PH stainless steels in
titanium, tantalum and zirconium. dilute sulphuric acid slurry. Journal of
Wear, 258: 693-700. iron and steel research International,
Burstein, G.T. and Sasaki, K. 2000. Effect of 13(5): 73-78.
Impact angle on the slurry erosion- Rajahram, S.S., Harvey, T.J. and Wood, R.J.K.
corrosion of 304L stainless steel. Wear, 2009. Erosion–corrosion resistance of
240: 80-94. engineering materials in various test
Hodgkiess, T., Neville, A. and Shrestha, S. conditions", Wear, 267: 244–254.
1999. Electrochemical and mechanical Sasaki, K. and Burstein, G.T. 1996. The
interactions during erosion–corrosion generation of surface roughness during
of a high-velocity oxy-fuel coating and slurry erosion-corrosion and its effect
a stainless steel. Wear, 233–235: 623– on the pitting potential. Journal of
634. Corrosion Science, 38(12): 2111-2120.
Hu, X. and Neville, A. 2005. The Stack, M.M. and Pungwiwat, N. 2004. Erosion-
electrochemical response of stainless corrosion mapping of Fe in aqueous
steels in liquid–solid impingement. slurries: some views on a new rationale
Wear, 258: 641–648. for defining the erosion-corrosion
Hutchings, I.M. 1992. Tribology: Friction and interaction. Wear, 256: 565-576.
Wear of Engineering Materials, Stack, M.M., Corlett, N. and Zhou, S. 1999.
Edward Arnold, London. Impact angle effects on the transition
Jana, B.D., Stack, M.M. 2005. Modelling boundaries of the aqueous erosion–
impact angle effects on erosion– corrosion map. Wear, 225–229: 190–
corrosion of pure metals: construction 198.
of materials performance maps. Wear, Toro, A., Sinatora, A., Tanaka, D.K. and
259: 243–255. Tschiptschin, A.P. 2001. Corrosion–
Lopez, D., Congote, J.P., Cano, J.R., Toro, A. erosion of nitrogen bearing martensitic
and Tschiptschin, A.P. 2005. Effect of stainless steels in seawater–quartz
particle velocity and impact angle on slurry. Wear, 25: 1257–1264.
the corrosion-erosion of AISI 304 and Wang, H.W. and Stack, M.M. 2000. The erosive
420 stainless steels. Wear, 259: 118- wear of mild and stainless steels under
124. controlled corrosion in alkaline slurries
Lopez, D., Falleiros, N.A., and Tschiptschin, containing alumina particles. Journal of
A.P. 2007. Corrosion–erosion behavior Materials Science, 35: 5263 – 5273.
192
Wharton, J.A. and Wood, R.J.K. 2004.

Influence of flow conditions on the
corrosion of AISI 304L stainless steel.
Wear, 256: 525–536.
Wood, R.J.K. 2006. Review erosion–corrosion
interactions and their effect on marine
and offshore materials. Wear, 261:
1012–1023.
Yao, Z., Zheng, Y. and Ke, W. 1995. The
influence of applied potential on the
erosion-corrosion behavior of AISI321
stainless steel in acidic slurry medium.
Wear, 186-187: 568-572.
193
INFLUENCE OF PLASMA SPRAY PARAMETERS ON WEAR PROPERTIES

OF TITANIA COATINGS
S. M. Forghani1, M. J. Ghazali1, A. Muchtar1, A. R. Daud2, N. H. N. Yusoff1, A.

Almaslow1
1
Department of Mechanical & Materials Engineering, Faculty of Engineering
National University of Malaysia
43600 UKM, Bangi, Selangor, Malaysia
E-mail: forghani@eng.ukm.my
2
School of Applied Physics, Faculty of Science & Technology
National University of Malaysia
43600 UKM, Bangi, Selangor, Malaysia.
ABSTRACT TiO2 coatings show that they improve wear

properties of the coatings (Yilmaz et al., 2007).
In this study the influence of various plasma Further study on anti-wear properties of titania
spraying parameters (plasma power, powder coatings are needed in order to develop their
feeding rate and scanning speed) on important applications in industries.
properties of titania (TiO2) coatings were
investigated. Fused and crushed (F&C) titania Atmospheric Plasma Spray (APS) is
(TiO2) powder particles (10-45 µm) were vastly applied to deposit ceramic coatings with
thermally sprayed via atmospheric plasma spray high melting temperatures. Properties of a
(APS) onto mild steel substrates. The plasma sprayed coating and its microstructure is
microstructure, microhardness, surface strongly dependent upon parameters such as
roughness and abrasive wear resistance of these spraying parameters and powder size (Lima and
coatings were analyzed and compared. The Marple, 2007). Parameters like plasma power,
Wear resistance of coatings was investigated powder feed rate, working gas composition and
applying pin-on-disk tests and microstructural scanning speed can affect coating properties
characterization was performed. The results (Lima and Marple, 2008). Applying the most
demonstrated that microstructure, wear rate and appropriate parameters for deposition of a
microhardness were profoundly affected by coating can optimize both properties and cost of
plasma power. It was found that higher plasma production. The aim of the current study is to
powers exhibited denser coating microstructures investigate the influence of plasma power,
and improved both microhardness and wear powder feed rate and scanning speed on
resistance. Surface roughness of the coatings microhardness, surface roughness and wear
was directly affected by powder feed rate and resistance of titania coatings on mild steel
inversely affected by scanning speed. substrates.
Keywords: Atmospheric Plasma Spray (APS), 2. METHODOLOGY

titania (TiO2), wear resistance
2.1 Experimental Design
1. INTRODUCTION
Full-factorial method helps studying of the
Titania (TiO2) coatings are developing as hard effects that several parameters may have on the
ceramic coatings for various purposes. (Ctibor results of a particular process. In 2-level full
et al., 2006). Their photocatalytic effects factorial design, responses are measured at all
(considering anatase phase) in destruction of combinations of the experimental parameter
organic pollutants has been recently under levels. A 23 factorial design for three
consideration (Chen et al., 2006). Their independent variables was used. Design of
biomedical applications have also been experiments was applied using Design Expert
investigated (Jaworski et al., 2010). These Software (statistics Made Easy, version 6.0.10,
coatings have fairly porous hard microstructures stat-Ease, inc., Minneapolis, MN). Plasma
which can be used under severe working Power (X1), powder feed rate (X2) and scanning
environments. Researchers on anti-wear Al2O3- speed (X3) were chosen as independent
194
parameters and the responses studied were the The powder composition was analyzed by X-
microhardness value (Y1), surface roughness Ray diffraction method (XRD) (Bruker, D8-
(Y2) wear rate (Y3). A central point was added Advance 2009, Germany). Microstructures of
to the design leading to a total of 9 sets of the coatings were characterized by scanning
experiments. Each response was modelled using electron microscopy (SEM). Hardness of the
the following equation: coatings was measured by a Vickers
microhardness tester (HMV-2T, Japan). A 300g
Yi = b0 + b1X1 + b2X2 + b3X3 + b12X1X2 + load was applied for 15 on samples cross
b13X1X3 + b23X2X3+ b123X1X2X3 (1) sections.
Abrasive wear tests were conducted

Where Yi is the theoretical response following the ASTM G-65 Method using a TR-
function (Pierlot et al., 2008). The independent 20LE pin on disc tester machine from Ducom.
variables, their experimental range and coded The samples were coated on 10 mm diameter
levels for the coating process are given in Table mild steel pins. The pin slid over a disc against a
1. SiC sand paper (120 grit). The applied load was
5 N. A new abrasive paper was applied for each
Table 1 Levels of independent variables test. The test was run for 20 meters. After the
test, samples were cleaned with ethanol
Range (coded level) ultrasonically and then dried in an oven for 15
Independent factors Low Central point High minutes at 50°C. The samples were weighed
(-1) (0) (+1) before and after the test to measure the mass
Plasma Power (X1) 20 30 40 loss. Abrasive wear rate was obtained dividing
KW the mass of the removed material by the load
Powder feed rate (X2) 6 14 22
g/min and the sliding distance. Surface roughness (Ra)
Scanning Speed (X3) 200 350 500 was measured by a SV-C3100 (Formtracer,
mm/s Mitutoyo) tester. The measurement was applied
on the coating surface along a length of 15mm
2.1 Materials and Coating Process with a pitch of 0.001 mm and at 2.0 mm/s.
Titania powder (fused and crushed Amperit 3. RESULT AND DISCUSSION

782.054, H.C. Starck, Germany) with a particle
size range of 10-45µm was used. Rutile was the 3.1 Phase Composition and Microstructure
main phase of the powder. Figure 1 illustrates
the morphology of the used powder. ASTM A- The XRD patterns of the powder and the
36 mild steel substrates (40×20×6 mm3) were coatings are shown in Figure 2. The powder is
grit blasted with alumina grits before coating almost rutile (99%). The XRD patterns do not
process. The deposition was performed by a show any distinctive differences in the phase
Praxair plasma gun (SG-100) mounted on a content of the coatings deposited. Rutile
robot. The stand-off distance was 100 mm. remains as the main phase in all samples. Rutile
is naturally a more stable phase than anatase.
From the mechanical view it is harder than
anatase and therefore having a nearly complete
rutile microstructure can possibly improve the
wear resistance of the coatings.
40KW
30KW
Intensity
20KW
R
Figure 1 Amperit 782.054 fused & crushed R R
R R
powder Powder A
A R
A
10 20 30 40 50 60
2θ°
2.2 Coating Characterization
195
Figure 2 XRD patterns for powder and coatings 3.3 Analysis of Variance
(R=Rutile, A=Anatase)
Table 2 shows the complete obtained responses
3.2. Microstructure for various factor level combinations:
The microstructure of plasma spray coatings is Table 2 Full-factorial 23 design for plasma spray
profoundly affected by plasma parameters. parameters
Figures 3 to 5 illustrate these changes in the
Run X X X Y Y Y
microstructure of the samples. It can be seen 1 2 3 1 2 3
that the content of unmelted particles and 1 -1 1 1 663 4.2260 0.0223
porosity is changed due to the changes in spray 2 1 1 1 778 4.2107 0.0159
parameters. These changes play an important 3 -1 1 -1 719 4.1667 0.0252
role in determining the mechanical properties of
4 -1 -1 1 615 3.7846 0.0197
the coatings. This is described with more details
in the next section. 5 1 -1 1 831 3.7493 0.0088
6 1 1 -1 798 4.5463 0.0111
7 1 -1 -1 878 4.0747 0.0067
8 -1 -1 -1 622 4.0933 0.0202
9 0 0 0 849 4.1000 0.0058
Regression models were developed

using Analysis of variance (ANOVA) to
recognize the effects of each parameter and its
interactions with the other parameters on each
of the responses. Sum of square (SS),
coefficients of the models, and P value for the
Figure 3 Microstructure of sample number 4 two factorial designs are presented in Table 3.
The P-value shows the probability of error
which was considered as a criterion for
accepting the factor in the model. P-values less
than 0.05 indicate model terms are significant.
Values greater than 0.10 indicate the model
terms are not significant. In other words, smaller
P-values make the model more significant.
Table 3 Regression coefficients and p-values for

the responses
Term SS* Coefficient P-value

Y1 65105.00 738 0.0004
X1 55444.50 83.25 0.0002
Figure 4 Microstructure of sample number 9
X1X2 9660.50 -34.75 0.0096
Y2 0.37 4.11 0.0169

X2 0.26 0.18 0.0122
X3 0.10 -0.11 0.0606
Y3 2.976E-4 0.016 0.0010

X1 2.520E-4 -5.613E-3 0.0005
X2 4.560E-5 2.387E-3 0.0197
*
Sum of Squares
Figure 5 Microstructure of sample number 7

3.4 Microhardness (HV0.3)
196
Equation (2) shows the relationship between a low surface roughness we should combine a
microhardness and plasma power and also its low feeding rate with a high scanning speed.
interaction with the amount of powder feeded.
Y1= + 738 + 83.25X1 – 34.75X1X2 (2)
This relation is illustrated in Figure 6.

It can be seen that to get a harder coating we
need to apply higher levels of plasma power and
combineit with low level of feeding rate. Table
3 shows that the role of plasma power is higher
that the feeding rate. A higher plasma power
results in a higher degree of melting which Figure 7 The Effect of powder feed rate and
causes better adhesion of the molten splats over scanning speed on surface roughness of the
each other during spraying and a denser coating coatings
with less porosities is achieved. In addition,
using a lower feeding rate will lead to a higher
percentage of the sprayed particles to get to
3.6 Wear rate
their melting point and the fraction of unmelted
particles presented in the coating is reduced.
Equation (4) derived from the regression
coefficients shows the effect of plasma power
and powder feed rate on wear rate of the
coatings. Like the microhardness, wear rate is
under the profound impact of plasma power.
Powder feed rate has a lighter effect.
Y3 = + 0.016 -5.613*10-3X1 + 2.387*10-3X2 (4)
To describe this phenomenon we have

to consider this fact that wear resistance of a
Figure 6 The Effect of plasma power and material is generally a function of its hardness
powder feed rate on microhardness of the and ductility. Ceramic materials show very low
coatings ductility levels which are negligible compared
to metals. In other words, a ceramic’s wear
resistance is in direct relation with its hardness.
3.5 Surface Roughness (Ra) As we mentioned earlier and in Eq. (2), the
hardness of the coatings was related to the
plasma power and also in a lower level to the
Surface roughness of the coatings is mainly
powder feed rate. As described in section 3.4,
affected by powder feed rate and scanning speed
higher plasma powers lead to denser and harder
(Eq. (3)).
coating structures. This means that the abrasive
particles in the wear test would be more
Y2 = + 4.11 + 0.18X2 – 0.11X3 (3) successful in detaching the layers of a softer and
less uniform coating deposited by a lower
The positive effect of the powder feed plasma power. Conversely, the powder feed rate
rate and negative effect of scanning speed is has a positive effect on wear rates. It means that
shown in Figure 7. When the powder feed rate an increase in the feed rate has a destructive
is high a large amount of particles are propelled effect on wear resistance of the coatings. For a
towards the substrate and the time for complete given plasma power, particles in a trajectory of
flattening of these particles would not be a high feed rate will absorb less thermal energy
enough. These particles would pile up and form and a smaller fraction of those particles can
a rougher surface. Applying a lower scanning attach properly to the surface. It means that the
speed will intensify this effect. In order to have percentage of semi-molten and unmolten
particles increases in the coating. Under wear
197
condition these particles can be removed more roughness. When the powder jet passes
easily leading to a higher rate of wear. This faster along the substrate a smaller
effect is illustrated in Figure 8 . In order to have number of particles is propelled to a
a lower wear rate we should apply higher certain area in a single pass and
plasma powers with lower powder feed rates. therefore less splat pile-up is occurred.
This results in lower levels of surface
roughness.
REFERENCES
CHEN, H., LEE, S. W., KIM, T. H. & HUR, B.

Y. 2006. Photocatalytic decomposition
of benzene with plasma sprayed TiO2-
based coatings on foamed aluminum.
Journal of the European Ceramic
Society, 26, 2231-2239.
CTIBOR, P., BOHÁČ, P., STRANYÁNEK, M.
& ČTVRTLÍK, R. 2006. Structure and
mechanical properties of plasma
Figure 8 The Effect of plasma power and sprayed coatings of titania and
powder feed rate on wear rate of the coatings alumina. Journal of the European
Ceramic Society, 26, 3509-3514.
JAWORSKI, R., PAWLOWSKI, L., PIERLOT,
4. CONCLUSION C., ROUDET, F., KOZERSKI, S. &
PETIT, F. 2010. Recent developments
Titania (TiO2) coatings were successfully in suspension plasma sprayed titanium
deposited on mild steel substrates using plasma oxide and hydroxyapatite coatings.
spray technique. The effects of plasma power, Journal of Thermal Spray Technology,
powder feed rate and scanning speed on 19, 240-247.
microhardness, surface roughness and wear LIMA, R. S. & MARPLE, B. R. 2007. Thermal
resistance of the coatings were investigated. The spray coatings engineered from
results are as follows: nanostructured ceramic agglomerated
powders for structural, thermal barrier
- Plasma power has a significant positive and biomedical applications: A review.
effect on microhardness and wear Journal of Thermal Spray Technology,
resistance of the coatings. This is due 16, 40-63.
to higher fractions of melted particles LIMA, R. S. & MARPLE, B. R. 2008. Process-
which makes a denser and more property-performance relationships for
homogeneous. A higher power titanium dioxide coatings engineered
improves the hardness of the coating from nanostructured and conventional
towards the bulk titania. powders. Materials and Design, 29,
1845-1855.
- Powder feed rate has a main effect on PIERLOT, C., PAWLOWSKI, L., BIGAN, M.
determining the surface roughness and & CHAGNON, P. 2008. Design of
wear rate of the coatings. Higher experiments in thermal spraying: A
feeding rates result in higher surface review. Surface and Coatings
roughness due to a faster pile-up of the Technology, 202, 4483-4490.
splats in a certain area. It also raises the YILMAZ, R., KURT, A. O., DEMIR, A. &
possibility of weak attachment of TATLI, Z. 2007. Effects of TiO2 on the
particles onto the substrate which eases mechanical properties of the Al2O3-
the removal of those weak splats while TiO2 plasma sprayed coating. Journal
the coating is under wear conditions. of the European Ceramic Society, 27,
1319-1323.
- Scanning speed is a main factor only

for determining the surface roughness
of the coatings. Applying higher
scanning speeds will reduce the surface
198
SPECIFIC WEAR RATE OF KENAF EPOXY COMPOSITE AND OIL PALM

EMPTY FRUIT BUNCH (OPEFB) EPOXY COMPOSITE IN DRY SLIDING
Salmiah Kasolang, Anizah Kalam and Mohamad Ali Ahmad
Faculty of Mechanical Engineering

Universiti Teknologi MARA
40450 Shah Alam, Selangor, Malaysia
E-mail: alie_76_02@yahoo.com
ABSTRACT From the tribological point of view, few works

This paper presents an experimental have been pursued on kenaf (Sastra et.al 2005),
investigation carried out to compare specific betelnut (Chin and Yousif, 2009; Umar et.al,
wear rate and surface morphology between two 2010), carbon (Li and Xia, 2010) and E-glass
types of natural fibres namely kenaf and oil (El-Tayeb and Gadelrab, 1996) fibres and
palm empty fruit bunch (OPEFB). Kenaf fibres promising results were reported on the use of
were received in long fiber size and OPEFB in kenaf as alternative. For instance, the presence
different sizes (100, 125, 180 and 250 µm). of kenaf fibres has enhanced the wear
Both materials were mixed with the epoxy resin performance of composites (Chin and Yousif,
to produce epoxy composites. Wear tests were 2009). In another work on epoxy matrix filled
carried out using Abrasion Resistance Tester in with hard powder, it was observed that wear rate
dry sliding condition. These tests were increases as the normal load increases (Crivelli
performed at room temperature for different et al., 2001). On Oil Palm Empty Fruit Bunch
loads and at a constant sliding velocity of (OPEFB) fibres, many of previous work on
1.4m/s. Based on the results, the specific wear morphology, chemical constituents and
rate of Kenaf Fibre composite starts to converge properties of OPEFB have been reviewed by
to one similar value beyond 6km distance. In the (Shinoj et al. 2011). However, it was noticed
case of OPEFB epoxy composite, it was found that many of the researchers have focused more
the fiber size of 100 um has produced the on the physical properties, mechanical
highest specific wear rate. properties, water absorption characteristics and
degradation/weathering effects (Rozman et.al.
2001; Rozman et al. 2004; Kalam et.al 2005).
Keywords: Abrasive wear, Thermoplastic, Literatures on the dynamic mechanical analysis,
Thermoset, Surface Morphology, Roughness. abrasion test, thermal conductivity and electrical
properties were limited.
1. INTRODUCTION Wear - it was defined as the loss of
material from one or both of contacting surfaces
Natural fiber usage in industry promotes a when subjected to relative motion, while a
sustainable material development through broader definition of wear include any form of
utilization of renewable resources. Composite surface damage caused by rubbing processes on
reinforcement with natural fibres becomes an one surface against another (Kasolang et. al.
interest to many researchers and designers 2011). Abrasive wear occurs when hard
because of the positive aspects of the product. particles or hard asperities rub against a surface
Nowadays, all of these natural fibres are to cause damage or material removal. From
beginning to find their way into commercial Archard wear model (Archard, 1953), a specific
applications such as in automotive industries, wear rate, Ws, can be generally determined by,
marine hardware and household applications
(Maleque and Belal, 2007). Ws= (∆m) / (L x ρ x F) (1)
From a mechanical point of view, natural
fibres are good substitutes for polymeric Ws in (mm3/Nm);
composites due to advantageous characteristics ∆m is weight loss (g);
in certain aspects such as renewable nature F is applied load (N),
resources, low weight, cost, density and easily ρ is density (g/mm3) and
available. Another advantage is on the ease of L is sliding distance (m).
modification chemically and mechanically
(Maleque and Belal, 2007; Hanafi et.al ,2002).
199
2. APPARATUS AND EXPERIMENTAL

PROCEDURE
2.1 Material
In current study, raw kenaf fibers were supplied

by Malaysian Agriculture Research and
Development Institute (MARDI). Kenaf fibres
was supplied in long fibre. The epoxy resin was
mixed with hardener by composition ratios of
3:1 (by weight) respectively. The kenaf fibre
weight fraction of the composite is 20%. In the
case of OPEFB, raw OPEFB fibers used in the
composite were supplied by Sabutek (M) and
the epoxy resin and hardener by Leco (M).
OPEFB fibres was supplied in 100, 125, 180
and 250 µm in sizes. As per manufacturer
(a)
recommendations, resin and hardener were
mixed in a ratio of 10:6 parts by volume and
was later cured for 8 hours at room temperature.
The fibres fraction of the composite is 20% by
weight.
2.2 Wear test
Abrasion Resistance Tester (TR-600) as shown

in Fig. 1(a) was used for the dry sliding wear
tests. Kenaf and OPEFB fibres composite
sample was attached to rotational disc and put in
contact with two rotating abrasive wheels made
of vitrified bonded silicon carbide. Before each
test, abrasive wheels were cleaned from any (b)
dust using a dry brush. The size of composite Figure 1. (a) Abrasion Resistance Tester (b)
sample is 122 mm in diameter and 5 mm thick. Schematic diagram of the abrasion
The schematic diagram of the abrasion test test apparatus.
apparatus is shown in Fig. 1(b).
Summary of the operational conditions was Table 1. Operating parameters in abrasion test
tabulated in Table 1. For each applied load, a
new disc sample was used. The amount of Parameters Experimental
weight loss for each sample was measured conditions
before and after a test at suitable intervals by Type of motion Unidirectional sliding
weighing the disc sample to an accuracy of Contact geometry Cylinder on flat
+0.0001g using a precision balance. Specific Applied load (N) 5, 20 and 30 (Kenaf)
wear rate at each operating condition was 10 and 30 (OPEFB)
determined using equation (1). Sliding speed (m/s) 1.4
The surface condition of each material sample Temperature (oC) 24.8 ± 1.0
was analyzed using an optical 3D surface Humidity (%) 65 ± 2
measurement device. From the 3D surface Sliding distance (m) 1000 to 10 000 at
measurements, details of the surface interval of 1000
morphology and the surface roughness were
obtained.
200
3. RESULT AND DISCUSSION
The results on the mass loss over distance for

both kenaf and OPEFB epoxy composite are
shown in Fig. 2 and Fig. 3 respectively.
Generally, Kenaf epoxy composites mass loss
are consistently higher compared to OPEFB
composites. For the 30N load, the mass loss
value of Kenaf epoxy is almost 6 times higher
compared to that of OPEFB epoxy composite at
10km distance. In the case of OPEFB
composite, at 30N load, it was clear that the Figure 3. Specific wear rate profiles for kenaf
mass loss was significantly higher for the epoxy composites over distance
smallest fibre size (100µm) examined. The
value recorded was more than 6g. At other fibre Specific Wear Rate of Kenaf and OPEFB Epoxy Composite Distance
2.50E-02
sizes, the mass loss values were relatively
closed around 2g.
Specific wear rate (mm3/Nm

2.00E-02
1.50E-02
1.00E-02
5.00E-03
0.00E+00
1 2 3 4 5 6 7 8 9 10
Distance (km)
OPEFB_100µm_10N OPEFB_125µm_10N OPEFB_180µm_10N OPEFB_250µm_10N

OPEFB_100µm_30N OPEFB_125µm_30N OPEFB_180µm_30N OPEFB_250µm_30N
Figure 4. Specific wear rate profiles for OPEFB

epoxy composites over distance
From the results obtained, in the case of OPEFB

Figure 2. Mass loss of Kenaf and OPEFB epoxy composite, it can be safely said that at
epoxy composite over distance higher load, a smaller fiber size (i.e. 100 um)
tend to get worn out faster. For the rest of the
From the mass loss values, the specific wear sizes examined, the specific wear rate is not
rate profiles were produced for both Kenaf and distinctively different. In the observation of
OPEFB composites using equation (1) as shown surface morphology, the surface roughness of
in Fig. 3 and Fig. 4 respectively. For kenaf the sample before and after test was measured.
composite, the specific wear rate profile over The average surface roughness values for
distance is almost similar in all cases examined. OPEFB composite were tabulated in Table 2.
It was noticed that all three cases produced the
same wear rate starting at 3km distance. The Table 2. Average surface roughness, Ra values
wear rate starts to converge after 6km distance. before and after 10km for 10N and 30N
This is believed due to the fiber arrangement in loads
the composite matrix. For the first 5 km
distance, the loss was suspected due to epoxy
matrix alone on the surface. The values of
specific wear rate were later become lower and
more consistent as the kenaf fibers being
exposed.
For the case of OPEFB composites, the specific
wear rate profiles obtained were different from
those of the Kenaf. At 10N load, the specific
wear rate profiles over distance were different
with no specific trend for 100 and 125µm. The
specific wear rate for different sizes of fibres at
30N load was pretty much constant beyond
3000m distance. This behaviour could be
contributed by the homogenity of the mixing of
fibers and epoxy.
(a)
201
under different sliding contact

conditions”, Wear, vol. 192: 112-117.
Hanafi Ismail, Edyham M.R. and Wirjosentono
B. 2002. Bamboo fibre filled natural
rubber composites:the effects of filler
loading and bonding agent. Polymer
testing, vol. 21:139-144.
Kalam, A., Sahari, B.B., Khalid, Y.A., Wong,
S.V., 2005. Fatigue behaviour of oil
palm fruit bunch fibre/epoxy and
carbon fibre/epoxy composites.
Composite Structures 71: 34–44.
(b) Kasolang, S., Ahmad, M. A., Ghazali, F.A. and
Figure 4. Surface morphology of OPEFB epoxy Azmi, A.M., 2011. Preliminary study
composite capture by 3D microscope of dry sliding wear in Kenaf Epoxy and
imaging at 5000X magnification (a) Carbon Epoxy composites. Applied
before test (b) after 10km distance Mechanics and Materials Vols. 52-54:
464-469
From the surface morphology, distribution of Li J. and Xia Y.C. 2010. Evaluation of
fibres orientation can be determined. However, tribological properties of carbon fiber-
the correlation of fiber orientation to the specific reinforced PA6 composites. Polymer
wear rate has been explored in this study. Composites, vol. 31, Issue 3:536-542.
Maleque M.A. and Belal F.Y 2007. Mechanical
4. CONCLUSION Properties Study of Pseudo-Stem
Banana Fiber Reinforced Epoxy
Based on the results, the specific wear rate Composite. The Arabian Journal for
profiles over distance for kenaf and OPEFB Science and Engineering, Volume 32,
epoxy composites were determined. It can be Number 2B.
concluded that:- Rozman, H.D., Ahmadhilmi, K.R. and
1. Generally, the specific wear rate for kenaf Abubakar, A., 2004. Polyurethane
(PU)-oil palm empty fruit bunch (EFB)
epoxy composite is higher at the early test
composites: The effect of EFBG
and start to converge to a consistent value reinforcement in mat form and
after 6km. isocyanate treatement on the
2. In the case of OPEFB epoxy composite, at mechanical properties. Polymer
30N load, the smallerst fiber size Testing 23: 559-565
examined (i.e. 100 um) produces the Rozman, H.D., Tay, G.S., Abubakar, A. and
Kumar, R.N., 2001. Tensile properties
highest specific wear rate. of oil palm empty fruit bunch-
3. Surface roughness was consistently polyurethane composites. European
reduced after 10km in all cases. Polymer Journal 37: 1759-1765
Sastra H.Y, Siregar J.P., Sapuan S.M., Leman
REFERENCES Z. and Hamdan M.M. 2005. Flexural
properties of Arenga pinnata fibre
Archard, J.F., 1953. Journal of Applied Physic. reinforced epoxy composites.
24: 981. American Journal of Applied Sciences,
Chin C.W. and Yousif B.F. 2009. Potential of (Special Issue): 21-24.
kenaf fibres as reinforcement for Shinoj, A., Visvanathan, R. Panigrahi, S. and
tribological applications. Wear, vol. Kochubabu, M., 2011. Oil palm fiber
267: 1550-1557. (OPF) and its composites: A review.
Crivelli Visconti I., Langella A. and Durante M. Industrial Crops and Products 33: 7-22
2001. The wear behaviour of Umar Nirmal, Yousif B.F. Dirk Rilling and
composite materials with epoxy filled Bevern P.V. 2010. Effect of betelnut
with hard powder. Applied Composite treatment and contact conditions on
Materials, vol. 8: 179-189. adhesive wear and frictional
El-Tayeb N.S. and Gadelrab R.M. 1996. performance of polyester composites.
Friction and wear properties of E-glass Wear, vol. 268: 1354-1370.
fiber reinforced epoxy composites Yousif B.F. , Umar Nirmal and Wong K.J.
2010. Three-body abrasion on wear
202
and frictional performance of treated

betelnut fibre reinforced epoxy (T-
BFRE) composite. Materials and
Design, vol. 31: 4514-4521.
203
WEAR PROPERTIES INVESTIGATION OF RAIL AND WHEEL MATERIALS USING

PIN-ON-DISK METHOD
M. Bin Sudin and Windarta
Mechanical Engineering Department, Universiti Teknologi PETRONAS,

Tronoh 31750, Perak, Malaysia
E-mail: mustasu@petronas.com.my
ABSTRACT rolling, and impact loads), speed, temperature,

type of counter body (solid, liquid, or gas), and
The contact between wheel flange and railway type of contact (single phase or multiphase, in
gauge is categorized under area contact. This which the phases involved can be liquid plus
research was aimed to determine wear rate and solid particles plus gas bubbles).
wear mechanisms of the rail and wheel material. Rail tracks also sometime called
As received rail material sample microstructure railway tracks or railroad tracks are the surface
were studied using an optical microscope. The structures that support and guide trains or other
wear tests were performed on Ducom rail-guided transportation vehicles. The railway
multispecimen test machine designed according material must strong enough to support the load.
to ASTM G99 standards. The applied loads The contact between a heavily loaded wheel and
selected were 40, 60, 80, 100 and 120 N with rail caused spots contact area (William, 1994).
common rotating speed of 0.15, 0.31, 0.47, 0.63, In attempting to predict the likely damage to
1.57 and 3.14 m/s. The results show that components, or their life under a given set of
hardness value decreases up to 25% and 37 % operating conditions, knowledge, or at least a
from base material for less than 50 μm thin realistic estimate, of the true stresses
layer after velocity of 1.57 m/s and 3.14 m/s experienced by the material is crucially
respectively. In steady state condition, the important. At the macroscopic level these might
depths of wears are 88.63 μm, 125.66 μm, represent the contact between the wheel and
263.01 μm, 395.67 μm and 451.37 μm for rail, while at the microscopic level one can think
applied loads of 40, 60, 80, 100 and 120 N, about modeling the contact between individual
respectively. The depths of wear due to velocity surface asperities on two opposing surfaces.
variations are 395.67, 465.72, 604.66, 722.15, In the past, many papers on rail wear
843.97 and 918.16 μm for velocity of 0.15, were published. Hegadekatte et al (2006) have
0.31, 0.47, 0.63, 1.57 and 3.14 m/s, respectively. predicted wear rate rail using pin on disk test.
Wear rate values of materials increases with The wear rates were globally for the whole
increasing load. SEM micrograph showed specimen, did not consider which contact
plastic deformation caused by abrasive wear. surfaces part. These experiments did not
Coefficient of friction has same trend with wear consider elastic foundation for contact pressure,
rate. Wear rate increase by increasing applied deformation shear effect in contact, contact
loads. area, and role of friction and stress distribution
in wear process.
Keywords: pin-on-disk, rail, wear, wheel.
Jia et al (2007) reported sliding wear
behavior of copper alloy contact wire against
1. INTRODUCTION
the copper-based strip used in railway systems.
The tests used variations electrical currents of 0,
The Wear is the erosion of material from a solid
surface by the action of another surface 7.5, 15, 22, 30, 40 and 50 A with maximum
(Telliskivi, 2004). It is about to surface voltage E was controlled at 6 V in lab
interactions and the removal of material from a environment. The disk was rotated at 300 rpm
surface as result of mechanical action. Wear can for 4 h, corresponding to a sliding velocity of 5
be defined as a process in which interaction of m/s (18 km/h) and sliding distance of 72 km.
the surfaces or bounding faces of a solid with its The normal load of contact was 45 N. The wear
working environment results in dimensional rate of copper alloy increases with the
loss of the solid, with or without loss of increasing of electrical current. Adhesive wear,
material. abrasive wear and electrical erosion wear are the
Aspects of environment with affect dominant wear mechanisms during the electrical
wear include loads (sliding, reciprocating, sliding wear processes. In this research Jia et al.
204
used constant loads and speed. This situation Different with the previous researches, Delprete
had been improved by Witaszek and Witaszek and Rosso (2009) proposed an instrument and a
(2007). They investigated wear on the railway methodology for monitoring and diagnosis of a
component using Timken wear testing machine. rail. Delprete used a simple transducer for
The tests used three different normal loads (63, measuring the vertical and lateral forces at the
108 and 153 N) and sliding speeds were 0.12, same time. The results showed that instrument
0.25, 0.36 and 0.49 m/s. The tests were can be used to forecast life assessment of the
performed for 7.38, 14.76 and 22.15 m. The rail and to monitor ballast mechanical behavior.
This instrument did not consider damage model
results indicate that the wear depends on such
on rail and not valid for high speeds and bigger
parameter as load, sliding distance and speed.
loads.
The result showed that the higher the speed, the
The objectives of the research are to
more decreasing the wear due to the appearance
study on tribology properties of railway
of the oxide films on the rubbing surfaces. The
material. Tribology properties include wear
films prevent metal to metal contacts and rate, wear mechanisms and coefficient of
adhesion. In 2008, Ozsarac and Aslanlar (2008) friction. Wear testing used 40, 60, 80, 100 and
investigated wear behaviors of wheel-rail 120 N applied loads with rotating speed of 100,
interface in water lubrication and dry friction 200, 300, 400, 1000, and 2000 rpm.
using pin-on-disk. The tests applied various load
of 10, 20, 30 and 40 N. The results found that
the friction coefficient is decreasing in wet 2. METHODOLOGY
sliding experiments however the weight and
volumetric loss of rail materials less than that of 2.1 Material Preparation
wheel samples. An abrasive wear failure was
observed in dry and wet friction conditions Wear tests were performed in Ducom
according to SEM micrographs of samples multispecimen test machine designed according
taken from wheel and rail. to ASTM G99 standards. Pin was chosen from
Lewis and Olofsson (2003) have rail material and disk was chosen from wheel
mapped rail wear transitions using twin disk and material. The chemical composition of material
is shown in Table 1.
pin-on-disk machines to relate the
Figure 1 shows the shape cut out disk specimen.
measurements taken from the field. The twin
These locations were selected to represent rail
disk tests show that three regimes (mild, severe,
gauge–wheel flange contact. Pin samples were
and catastrophic) are presence. Using same
prepared as 6 mm in diameter and 12 mm in
machine, Vasauskas et al (2005) tested strength length. Disk samples were prepared as 40 mm in
of railway wheels under contact load. The diameter and 5 mm in width. Both pin and disk
results indicate existing anisotropy, in term of sample were polished and cleaned with alcohol
position of the test specimen. Material strength and dried.
is importance, but it is unclear which material
parameters that correlated to the resistance Table 1. Nominal Chemical Composition of
against subsurface cracks. the Studied Steel
Hegadekatte et al (2008) predicted C Mn P S Si Fe
modeling scheme for wear in tribometer. These Rail 0.71 0.87 0.02 0.01 0.29 Balance
researches used an approach that involves Wheel 0.53 0.69 0.02 0.01 0.28 Balance
computationally incremental implementation of
Archard’s wear model. These numerical
researches can be used to identify the wear 2.2 Microstructure Testing
coefficient from pin-on-disk experimental data
and also predict the wear depths within a limited Specimens were polished by using 120, 220,
range of parameter variation. The results 500, 800, 1000 and 1200 grit abrasive papers
and 1 μm diamond.
showed that Archard’s wear model have a good
agreement with the experimental data including
the case when the tribometer was lubricated
with water. Limitation of the research was a
constant average pressure assumed over the
contact area in any sliding distance increment.
The worn out surface was assumed to be always
flat so that contact area can be easily estimated.
205
Location of
disk specimen
Figure 2. Rail steel microstructure after etching

using 3 % nital
Figure 1. Location of disk specimen.
420
Thereafter, specimens were etched by 3 % nital
400
solution. Microstructure tests were conducted
380
using Leica DM LM.
360
Microhardness, VHN
340
2.3. Wear Testing
320
Both pin and disk samples used similar rail 300
material. The calculations of wear rate was done 280 1.57 m/s measured
for pin and disk materials in dry conditions. The 260
3.14 m/s measured
1.57 m/s prediction
testing was performed under rotating speed of 240
3.14 m/s prediction
100, 200, 300, 400, 1000, and 2000 rpm. Each 20
specimen was tested under 40, 60, 80, 100 and 0 20 40 60 80 100 120 140 160
120 N loads. Samples were weighted before and Distance away from worn surface, mm
after each test and weight losses were recorded.
Figure 3. Micro hardness measurements

3. RESULT AND DISCUSSION performed on railhead
3.1 Microstructure and Microhardness

Testing 3.2 Wear Testing
Figure 2 shows that the microstructure of rail In wear test, all of specimen were polished
sample is not homogenous. It is much like using 120, 220, 500 grit abrasive paper.
lamellar pearlite (dark) and ferrite (bright) Experiments were conducted in 100 RPM
structures. Ferrite structures have low micro rotating speed and applied contact loads were
hardness and a hardness profile also shows that 40, 60, 80, 100 and 120 N.
it has small variation between high and low Figure 4 showed the raw results
vales of hardness. Figure 3 shows micro obtained from wear test machine. In steady state
hardness decreases with distance from the worn condition, the depths of wears are 88.63 μm,
surfaces (Zhang et al, 2006 and Baumann et al., 125.66 μm, 263.01 μm, 395.67 μm and 451.37
1996). μm for applied loads of 40, 60, 80, 100 and 120
Micro hardness profile was taken from cross- N, respectively. In general, the Depth of wear
sectional surface under worn track. The results increases linearly with increasing applied
show hardness value decreases up to 25% and contact loads. Wear rate increases linearly in
37 % from base material for less than 50 μm each case during transient condition. These
thin layer after velocity of 1.57 m/s and 3.14 results have the same trend comparing to the
m/s respectively. studies conducted by Lewis & Olofsson (2003).
206
a
Figure 4. Depth of wear materials
c
Figure 5. Weight loss values of materials
The change of depth of wear due to velocity

variation is presented in Figure 5. The depths of
wear due to velocity variations are 395.67,
465.72, 604.66, 722.15, 843.97 and 918.16 μm
for velocity of 0.15, 0.31, 0.47, 0.63, 1.57 and
3.14 m/s, respectively. The wear rate values of
materials increases with increasing applied load
and velocity were obtained. This similar finding
was also supported by Telliskivi & Olofsson d
(2004) results.
Figure 6. SEM micrograph wear debris after
100 N applied load pin-on-disk tests (a) 0.62
m/s (400 rpm) pin material (b) 3.14 m/s (2000
rpm) pin (c) 0.62 m/s (400 rpm) disk and (d)
3.14 m/s (2000 rpm) disk .
207
The FESEM micrograph of pin and disk below the worn track. These values were
materials was shown in Figure 6. Wear scar obtained from test velocity of 1.57 m/s and 3.14
shows the present of third body particle on both m/s respectively. The steady state depths of
pin and disk surfaces which explain the abrasive wears on constant velocity tests are 88.63 μm,
wear mechanism has occurred at the early stage 125.66 μm, 263.01 μm, 395.67 μm and 451.37
of sliding. Due to the plastic deformation at the μm for applied loads of 40, 60, 80, 100 and 120
proceeding stage of wear interaction the N, respectively. The depths of wear due to
following severe adhesive wear mechanism has velocity variations are 395.67, 465.72, 604.66,
developed (Viafara et al., 2005). 722.15, 843.97 and 918.16 μm for velocity of
0.15, 0.31, 0.47, 0.63, 1.57 and 3.14 m/s,
respectively. Wear rate values of materials
increases with increasing load. SEM micrograph
showed plastic deformation caused by abrasive
wear. Coefficient of friction has same trend with
wear rate.
ACKNOWLEDGEMENT
The authors would like to thank Keretapi Tanah

Melayu Berhad (KTMB) for providing railway
and wheel materials. This research is under
Graduate Assistantship Universiti Teknologi
PETRONAS
Figure 7. Coefficient of Friction vs. sliding
distance
REFERENCES
0.7
0.6 Baumann, G., Fecht, H.J., Liebelt, S., 1996,

Formation of White Etching Layers on
0.5 Rail Treads, Wear, 191, pp133-140
Coefficient of Friction
0.4 Delprete, C. and Rosso, C., An Easy

Instrument and a Methodology for the
0.3 Monitoring and the Diagnosis of a
Rail, Mechanical Systems and Signal
0.2
Measured
Processing, vol 23, 2009, pp940-956.
Matsumoto et al Hegadekatte, V., Huber, N., Kraft, O.,
0.1
Lee & Polycarpou Modelling and Simulation of Dry
0 Sliding Wear in a Tribometer, IJTC,
2006.
0 20 40 60 80 100
Sliding Distance, m Hegadekatte, V., Kurzenhauser, S., Huber, N.,
Kraft, O., A Predictive Modelling
Figure 8. Comparison Coefficient of Friction Scheme for Wear in Tribometers, J
Tribology International, Vol 41, 2008,
pp1020-1031.
Coefficient of friction both pin and disk material Jia, S.G., Liu, P., Ren, F.Z., Tian, B.H., Zheng,
was shown in Figure 7. Coefficient of friction M.S., Zhou, G.S., Sliding Wear
Behavior of Copper Alloy Contact
increases with increasing sliding distance. Wire Against Copper-based Strip for
Coefficient of friction increase with increasing High-speed Electrified Railways,
applied load. This similar result (Figure 8) was Wear, Vol 262, 2007, pp 772-777.
supported by (Matsumoto et al., 2005, and Lee Lee, K.M., & Polycarpou, A.A, 2005, Wear of
and Polycarpou, 2005). Conventional Pearlitic and Improved
Bainitic Rail Steels, Wear, 259, pp391-
399
Lewis, R., and Olofsson, U., 2003, Mapping
Rail Wear Transitions, 6th
4. CONCLUSION International Conference on Contact
Mechanics and Wear of Rail/Wheel
In this study, wear characteristic for railway Systems.
material were investigated using pin-on-disk. Matsumoto, K., Suda, Y., Komine, H., Nakai,
Hardness value decreases down to 25% and 37 T., Tomeoka, M., Shimizu, K.,
Tanimoto, M., Kishimoto, Y., Fujii, T.,
% from base material starting from 50 μm 2005, A Proposal of Wheel/Rail
208
Contact Model for Friction Control, Viafara, C.C., Castro, M.I., Velez, J.M., Toro,
Journal of Mechanical Science and A., 2005, Unlubricated Sliding Wear of
Technology, Vol. 19 No. 1, pp437 – Pearlitic and Bainitic Steels, Wear,
443 259, pp405-411
Ozsarac, U. and Aslanlar, S, 2008, Wear Williams, J.A., 1994, Engineering Tribology,
Behaviour Investigation of Wheel/rail Oxford Univ. Press
Interface in Water Lubrication and Dry Witaszek, M., and Witaszek, K., 2007,
Friction, Industrial Lubrication and Laboratory Wear Assessment of
Tribology, Vol 60 no. 2, pp101-107. Selected Elements of Railway
Telliskivi, T., & Olofsson, U., 2004, Wheel-Rail Transport Means, Transport Problems,
Wear Simulation, Wear, 257, pp1145- pp71-77.
1153 Zhang, H.W., Ohsaki, S., Mitao, S., Ohnuma,
Telliskivi, T., 2004, Simulation of Wear in a M., Hono, K., 2006, Microstructural
Rolling-Sliding Contact by a Semi- Investigation of White Etching Layer
Winkler Model and the Archard’s on Pearlite Steel Rail, Materials
Wear Law, Wear, 256, pp817-831 Science & Engineering A, 421, pp191-
Vasauskas, V., Bazaras, Z., Capas, V., 2005, 199
Strength Anisotropy of Railway
Wheels Under Contact Load,
Mechanika, Nr. 1 (51).
209
EFFECTS OF MANGANESE ADDITION ON HARDNESS OF AA6063 ALLOY

FOR WEAR RESISTANT MATERIAL APPLICATIONS
H. Pratikno1,3, M.J. Ghazali1 and A.R. Daud2
1
Department of Mechanical and Materials Engineering, Faculty of Engineering and Built Environment,
Universiti Kebangsaan Malaysia, Bangi, 43600, Selangor, Malaysia
E-mail: hermanp@vlsi.eng.ukm.my, mariyam@eng.ukm.my
2
School of Applied Physics, Faculty of Science and Technology, Universiti Kebangsaan Malaysia,
Bangi, 43600, Selangor, Malaysia
E-mail: ard@ukm.my
3
Department of Ocean Engineering, Faculty of Marine Technology ITS,
Campus ITS Sukolilo, Surabaya, 60111, Indonesia
E-mail: hermanp@oe.its.ac.id,
ABSTRACT increased gradually and linearly as the

percentage of Mn was increased due to possible
AA6063 alloy is usually used in boat and ship solid solution hardening in the Al-matrix
building. This is because of its interesting (Seifeddine et al. 2008).
properties such as lightweight and aesthetic
characteristics. In this work, the AA6063 alloy Generally, metal alloys of high hardness
was melted together with 0.5 to 2.5 wt% of Mn will exhibit good wear resistance. Jeong et al.
and cast. The alloy was solutionised at 535 oC (2002) found that the abrasive wear resistance
for 6 hours and followed by water-quenching. of some annealed metals and steels was linearly
Artificial aging was performed on the alloy at proportional to their hardness. Annealed 5XXX
200 oC for 5 hours and followed by natural aluminium alloy has shown a positive
aging for 14 days. Vickers micro-hardness correlation between abrasive wear resistance
values indicated that the addition of Mn has and subsurface hardness (Mezlini et al. 2004).
increased hardness of the AA6063 alloy. A Dry sliding wear resistance increased with
combination of artificial and natural aging has increasing hardness for 7075 aluminium alloy
increased hardness of the alloy containing 2.5 after subjecting to retrogression and re-aging
wt% Mn from 120 VHN to 160.09 VHN. treatment (Baydogan et al. 2004).
Therefore, the increase in its hardness could
lead to the improvement of its wear resistance. In marine environment, corrosion-
resistant architectural Al-Alloy i.e. AA6063
Keywords: AA6063 alloy, manganese, hardness alloy is the most popular (Kasten, 1997). The
presence of magnesium and silicon has made
the alloy heat treatable and weldable with good
1. INTRODUCTION mechanical properties. It also has good surface
finish, high corrosion resistance and can be
Manganese (Mn) has been known to be an easily anodised. American Bureau of Shipping
important alloying element in aluminum alloys (ABS) and the American Society for Testing
which contributes to a uniform deformation. It and Materials had confirmed the AA6063 alloys
was reported that aluminium alloys with more with solution heat treatment and artificial aging
than 0.5wt% of Mn content like 6000 and 7000 (T6) are suitable for ship building (Ferraris,
series alloys, both showed a significant increase 2005).
in the yield and ultimate tensile strength without The focus of this study is to investigate
decreasing any ductility (Nam & Lee, 2000). It the effect of manganese addition in AA6063
was reported that the strength of alloys alloy particularly on the hardness properties.
210
2.4 Hardness Test
2. METHODOLOGY Prior to testing, the samples were mechanically

ground using SiC papers and polished with
2.1 Experimental Design diamond spray to 6 micron surface finish.
Hardness test was carried out by using a Micro-
In this work, the dependence of microhardness Vickers hardness tester, with a load of 300 g for
of AA6063 alloy on manganese content and 15 sec indentation time (ASTM, 2008). The
heat treatment temperature was investigated. reported microhardness values were an average
Response surface methodology (RSM) was used value of 15 measurements.
to model the correlation amongst the above
parameters.
2.5 Microstructural Study
Response surface methodology (RSM) is
a collection of mathematical and statistical The mechanical properties of alloys including
techniques. It is useful for modeling and hardness are strongly influenced by their
analysis of problems in which a response of microstructure. For microstructural study, the
interest is normally influenced by several samples were etched in Keller’s reagent and
variables. observed under an optical microscope.
In this work, the response surface

methodology (RSM) used a 3-Level Factorial 3. RESULTS AND DISCUSSION
where two numeric factors were applied,
namely, addition of Mn(wt%) and heat Figure 1 shows that microhardness of
treatment temperature(oC) with an assumption AA6063 alloy increases as the Mn content is
that the effect of solutionisation on hardness of increased but decreases when heat-treatment
artificially aged samples is constant. The lowest temperature is decreased.
composition of Mn was taken as 0% wt and the
highest value was 2.5%wt. The lowest heat DESIGN-EXPERT Plot
treatment temperature was 27oC (natural aging hardness

X = A: Mn
temperature) and the highest temperature was Y = B: Treatment
535oC(solution treatment temperature). The
160.09
lowest and the highest hardness values for those
140.08
heat-treatment temperatures were used to
Microhardness 120.07
construct the model. The response of the RSM (VHN)
100.06
was microhardness(VHN).
80.05
2.2 Material Preparation
AA6063 aluminum alloy was used as the 535.00

2.50
starting material. The chemical composition of 408.00
1.88
the AA6063 alloy in weight percent is 0.53 Si, Temperatur Treatment 281.00
1.25
(oC )
0.44 Mg, 0.28 Zn, 0.25 Fe, 0.03 Ti, 0.02 Cu, 154.00 0.63 Mn ( %wt )
0.01 Cr, 0.01 Co and Al balance. Mn was added 27.00 0.00
to AA6063 alloy and melted at 800oC with

holding time of 10 minutes. The melt was cast
in a steel mould with dimension of 14 mm in
diameter and 120 mm in length. The Mn content Figure 1. A plot indicating the dependence of
of the alloy was varied from 0.5 to 2.5 wt%. microhardness of AA6063 on Mn content and
heat-treatment temperature obtained using RSM
2.3 Heat Treatment 3-Level Factorial
Cast samples were then solution heat-treated at

535oC for 6 hours and water-quenched to room The variation of hardness values of the
temperature. This was followed by an artificial as-cast alloy and the alloys which were
aging at 200oC for 5 hours (Chen et al. 2000). subjected to solution treatment, artificial aging
The samples were then naturally aged for 14 and natural aging is shown in Figure 2. The
days (Daud & Wong, 2004). hardness of as-cast AA6063 alloy without Mn is
66.63 VHN and increased to 101.75 VHN with
211
2.5 wt % Mn (Figure 2). The increase is due to rather bigger grain size than that of the alloy
the decrease in grain size (Figure 3). containing 2.5 wt% Mn.
Solution treatment had increased the

220 as cast microhardnes of the alloys containing Mn of
solution treatment
less than 2.5 wt% which is due to a better
200 distribution of Mn compared to the as-cast alloy
solution treatment + artificial aging (Figure 4). Solution treatment seemed to
180 solution treatment + artificial & natural improve distribution of Mn, especially for the
Microhardness ( VHN )
aging alloys containing Mn of less than 2.5 wt%. The

160 hardness of the as-cast and solutionised alloys
with 2.5 wt% Mn was almost the same since
140 there is no significant different in Mn
distribution in both of them (Figure 4).
120
After solution treatment and artificial
100 aging, the hardness of this alloy was found to
increase from 93.76 VHN without Mn, to
80 150.42 VHN when containing 2.5 wt % Mn.
The hardness of the alloy containing 2.5 wt%
60 Mn has increased further up to 160.09 VHN
after artificially aged solutionised alloys was
40 subjected to 14 days natural aging.
0.0 0.5 1.0 1.5 2.0 2.5
Artificial aging has refined the grains with
Mn ( wt % )
better defined grain boundaries (Figure 5).
Furthermore, when the AA 6063alloys were
Figure 2. Effects of Mn additions on the allowed to undergo natural aging for 14 days,
microhardness of AA6063 alloy the alloys retained fine grains but fine Mn
precipitates and perhaps intermetallic phases
Figure 3 exhibits the microstructures of seemed to be uniformly distributed in the alloys
the as-cast alloy with 0.5 and 2.5 wt % Mn. The (Hwang et al, 2008), as shown in Figure 6.
as-cast alloy containing 0.5 wt % Mn has a
(a) (b)
Figure 3. Microstructure of as-cast AA6063 alloys with Mn (a) 0.5 wt % and (b) 2.5 wt %.
212
(a) (b)
Figure 4. Microstructure of AA6063 alloys with Mn (a) 0.5 wt % and (b) 2.5 wt % after solution treatment
(a) (b)
Figure 5. Microstructure of AA6063 alloys with Mn (a) 0.5 wt % and (b) 2.5 wt % after artificial aging.
(a) (b)
Figure 6. Microstructure of AA6063 alloys with Mn (a) 0.5 wt % and

(b) 2.5 wt % after artificial and natural aging
213
The well distributed precipitates and fine technologies, perspectives, In:

grains have improved further the hardness of the Proceeding of The 5th International
AA6063 alloys containing Mn where the Forum on Aluminium Ships, Tokyo,
hardness reached 160.09 VHN for the alloy Japan.
containing 2.5 wt% Mn. The increase in
Hwang, J.Y., Doty, H.W. and Kaufman, M.J.,
hardness which is related to grain refinement
2008. The effect of Mn additions on the
and the formation of precipitates indicates that
microstructure and mechanical
the alloy may have a better wear resistance
properties of Al-Si-Cu casting alloys,
(Mezlini et al. 2004, Baydogan et al. 2004).
Materials Science Engineering A488,
496-504.
4. CONCLUSIONS Jeong, D.H., Erb, U., Aust, K.T. and Palumbo,
G, 2002, The relationship between
The hardness of AA6063 alloy increases with hardness and abrasive wear resistance of
the Mn content. Both artificial and natural aging electrodeposited nanocrystalline Ni-P
processes have further improved the hardness coatings, Scripta Materialia 48, 1067-
property of the alloy. The increase in hardness is 1072.
due to the formation of the precipitates and
Kasten, M., 1997. Marine Metals Reference,
grain refinement. An addition of 2.5 wt% of Mn
to AA6063 alloy followed by artificial and 14 Metal Boat Quarterly, Washington
days natural aging has increased the hardness of Mezlini, S., Kapsa, Ph., Abry, J.C., Henon, C
this alloy to 160.09 VHN. The increase of the and Guillemenet, J. 2004, Abrasion of
alloy hardness, hence a better wear resistance aluminium alloy : effect of subsurface
can be achieved. hardness and scratch interaction
simulation, Wear 257, 892-900.
ACKNOWLEDGEMENT Nam, S. W. and Lee, D. H., 2000. The effects

of Mn on the mechanical behavior of Al
The authors would like to acknowledge alloys, Metals and Materials
Universiti Kebangsaan Malaysia for financial International, 6 (1), 13-16.
support through Research Grant UKM-DLP- Seifeddine, S., Johansson, S, and Svensson, I.L.,
2011-012. One of the author (Mr. Pratikno) 2008. The influence of cooling rate and
would also like to thank the Ministry of manganese content on the β-Al5FeSi
National Education of the Republic of Indonesia phase formation and mechanical
(DIKTI) for the financial support. properties of Al-Si based alloys.
Materials Science & Engineering A, 490,
REFERENCES 385 -390.
ASTM Section III., 2008. Metal Test Methods

and Analytical Procedures. ASTM
International, New York.
Baydogan, M., Cimenoglu, H. and Kayali, E.S.,
2004. A study on sliding wear of a 7075
aluminium alloy, Wear 257, 852-861
Chen, R., Iwabuchi, A. and Shimizu, T., 2000.
The effect of a T6 heat treatment on the
fretting wear of a SiC particle-reinforced
A356 aluminium alloy matrix composite,
Wear 238, 110-119.
Daud, A.R. and Wong, K.M.C., 2004. The
effect of cerium additions on dent
resistance of Al-0.5Mg-1.2Si-0.25Fe
alloy for automotive body sheets,
Materials Letters 58, 2545-2547.
Ferraris, S., 2005. Aluminium alloys in third
millennium shipbuilding : materials,
214
FRICTION AND WEAR CHARACTERISTICS OF PALM BIODIESEL
1
M. A. Fazal, 2A. S. M. A. Haseeb*, 3H. H. Masjuki
1-3
Centre for Energy Sciences, Faculty of Engineering, University of Malaya,
50603 Kuala Lumpur, Malaysia
e-mail: 1jewel_mme.buet@yahoo.com, 2haseeb@um.edu.my, 3masjuki@um.edu.my
ABSTRACT removal of the sulfur-containing compounds

rather than the reduction of polar compounds
Biodiesel, as an alternative fuel, is getting more containing other heteroatoms such as oxygen
acceptances in automotive applications. The and nitrogen (Dimitrakis, 2003). Consequently,
tribological behavior of biodiesel is crucial for the lubricity of diesel fuels is decreased after
its application. In the present study, friction and being modified for the safety of environment.
wear characteristics of palm biodiesel were Apart from this, rapid growth of modernization
investigated by using four-ball wear machine. and technological developments are leading
Biodiesel (B100), diesel (B0) and three different towards the depletion of limited fossil fuel
biodiesel blends like B10, B20, B50 were resources of the world. Therefore, it is very
investigated. Tests were conducted at 75ºC important to develop such an alternative fuel
under a normal load of 40kg for 1h at four which can meet these growing concerns.
different speeds viz, 600, 900, 1200 and 1500
rpm. Worn surfaces of the balls were Increasing qualitative requirements of the fuels
investigated by SEM. Results showed that wear of modern diesel engine can be satisfied a lot by
and friction were decreased with increasing the using biodiesel or its blends with diesel. It is
concentration of biodiesel. one of the most promising alternative fuels to
meet the required fuel properties and to protect
Key Words: Four-ball, Wear and friction, Palm the engine and environment (Demirbas, 2009).
biodiesel. It can be used in diesel engine with little or no
modification. It also offers some technical
1. INTRODUCTION advantages as compared to conventional diesel
fuel. The most common advantages of biodiesel
For engine fuel, lubricity is essential to prolong over petroleum diesel are derivation from a
the machine life through providing protection to renewable feedstock and domestic origin,
the moving surfaces against wear. Good biodegradability, higher flash point, reduced
lubricity is also important to cut down the exhausted emission and toxicity (Ozsezen and
energy consumption by reducing friction of Canakci, 2011; Fazal et al. 2011). Although
automotive parts (Tung and McMillan, 2004). these attributes of biodiesel have drawn
For some specific components of automobile attention as an alternative fuel, there exist some
such as fuel injectors and pumps, the lubricity significant drawbacks which have limited its
issue is very important as they are lubricated by commercial applications. These include auto-
the fuel itself (Celik and Aydin, 2011). The oxidation, corrosive nature, lower volatility,
temperature of the fuel contacted with these reactivity of unsaturated hydrocarbon chains
components is appeared to be over 60˚C (Has- (Gan and Kiat, 2010; Haseeb et al. 2010a)
imoglu et al., 2008) which can also influence which are more prone to influence the
the lubricity. In fact, the lubrication function for lubricating parameters, wear of various engine
diesel fuel is provided by its natural components components and so on. The impact of these
such as sulfur, aromatics and other heterocyclic, issues can also be varied for different
polar compounds (Sperring and Nowell, 2005). concentration of biodiesel in diesel blends.
However, due to environmental protection and Besides, the adaptation of a selected alternative
human health reasons, the sulfur content of fuel to suit diesel engine is considered more
diesel fuels is reduced to 50 ppm in the economically attractive in using it as blend with
European Union beginning with 2005 and the diesel fuel than in its pure form (Fraer et al.
limit is further tightened (≤ 10 ppm) for 2005- 2005). Several laboratory works have reported
08 (Shiroyama, 2007). The reason for the poor better performance of biodiesel as an additive in
lubricity of low-sulfur petrodiesel is not the lubricating oil (Maleque et al., 2000; Masjuki
215
and Maleque, 1997). Such effectiveness was balls were then investigated by using scanning
also reported for even lower (<1% biodiesel) electron microscopy.
blend levels (Karonis et al., 1999; Anastopoulos
et al., 2001). At the same time, some inferior
properties of biodiesel have also been reported.
Masjuki et al. (1997) found that above 5% palm
oil methyl ester (POME) in lubricant caused
oxidation and corrosion. According to Maleque
et al., (2000) for 5% POME in lubricant, the
4
total acid number (TAN) increases at 1
temperatures above 80˚C. They found that at
higher temperature (above 80˚C), oxidation of
the lubricants caused increased wear. These 2
results seem to imply that lubrication properties 5
of biodiesel can be varied depending on its 3
concentration in blend as well as test condition.
The present study aims to characterize the
lubricity in terms of friction and wear for
different concentration of biodiesel in blends as
Figure 1: Schematic diagram of four ball wear
compared to that of diesel fuel.
geometry: 1-rotating gripper for upper ball, 2-
test fuel, 3-cup for griping stationary three balls,
2. EXPERIMENTAL
4-rotating single ball, 5-stationary ball.
Friction and wear characteristics of palm oil Table 1. Conditions for the four-ball wear test
methyl ester were investigated by four ball wear
Test parameters
machine, IP 239/85. The test parameters in the
Applied load (kg) 40
present study were regulated by ASTM D4172
Rotation (rpm) 600, 900, 1200, 1500
standard except the speed which was 600 –
Fuel temperature (ºC) 75
1500 rpm instead of 1200 rpm. Details of the
Test duration (s) 3600
test conditions are described in Table 1. Palm
Test Ball
oil methyl ester used in this study meets
Materials Chrome alloy steel
EN14214 specifications and was supplied by
Composition C: (0.95-1.10)%, Cr:
Golden Hope Biodiesel Sdn Bhd, Malaysia. The
(1.3-1.6)%, Fe: balance
analysis report provided by the supplier is
Diameter 12.7 mm
summarized elsewhere (Haseeb et al. 2010b). In
Hardness 62 HRc
addition to pure diesel (B0) and biodiesel
Surface roughness (Ra) 0.040 μm
(B100), three different blends such as B10 (10%
biodiesel in diesel), B20, B50 were made on
volume basis for investigating the lubricity. 3. RESULTS
The schematic diagram of four-ball wear The variation of friction coefficient, calculated
machine is seen in Figure 1. Among the four form recorded torque has two parts, one is run-
balls, the lower three were held in fixed position in period and another one is steady state. Figure
against each other in a steel cup by means of a 2 shows that at the very beginning of each test
clamping ring. Another ball into the upper (during run-in period), the friction coefficient
chuck was rotating one. Testing fuel was poured was unstable with time and few minutes later it
in sufficient amount (approx 10 ml) to cover the came to a stable condition.
balls to a depth of at least 3 mm. During each
test, friction torque was recorded in order to
calculate the friction coefficient. The wear scar
diameters were measured before removing the
balls from the cup. The results reported here are
mean wear scar diameter of three balls. For
removing the worn products, the worn surfaces
were scrubbed lightly in a stream of water with
polymer brush so as not to mechanically abrade
the original surface. These were then degreased
with acetone. The cleaned worn surfaces of
216
coefficient (FC) in different fuels for last 1000s

of each test.
Figure 5 shows that the average friction

coefficient decreases with increasing the
concentration of biodiesel. For pure diesel and
biodiesel, the increasing rates of average friction
coefficient with speeds are almost linear while
for blends are not.
B0 B10 B20
B50 B100
0.09
Figure 2: Variation of friction coefficient with
respect to time (for first 1500s) at speeds (a) 600
Friction Coefficient
0.0885
rpm and (b) 1500 rpm.
Figure 3 shows that run-in period decreases with 0.087

increasing the concentration of biodiesel.
Decreasing trend of run-in period is faster for
0.0855
the blend containing up to 20% biodiesel and
beyond that concentration it becomes very slow.
However, it is seen that the overall run-in period 0.084
decreases with decreasing speed. 600 900 1200 1500
Speed (rpm)
Figure 5: Effect of speed on average friction

coefficient in different fuels under a fixed load
at temperature 75 ºC for 1 h.
It is seen in Figure 6 that the wear rate increases

with increasing the speed. On the other hand,
with increasing biodiesel concentration, wear
rate is decreased. The increasing trends of wear
rates follow the increasing trends of friction
coefficient as shown in Figure 5.
Figure 3: Fall in run-in period with increasing B0 B10 B20
biodiesel concentration in blends as well as with 1.02 B50 B100
decreasing speed.
0.89
Steady state friction coefficient for the last
WSD (mm)
1000s of each test was calculated from the

0.76
recorded torque. It is seen in Figure 4 that at
higher percentage of biodiesel (>20%), the
steady state friction coefficients are almost the 0.63
same for all speeds and concentrations of
biodiesel at about 0.083FC.
0.5
600 900 1200 1500
Speed (rpm)
Figure 6: Effect of speed on Wear Scar

Diameter in different fuels under a fixed load of
40 kg at temperature 75 ºC for 1h.
Figure 7 shows the SEM picture of worn

surfaces of the balls. Figures 7(a) to (d)
demonstrate that wear scar diameter decreases
with increasing the concentration of biodiesel.
Deformation of surface is also decreased with
increasing biodiesel. Abrasive wear is found for
Figure 4: Change in steady state friction
217
the sample in B100 while for B0, the wear can reported that the formation of several
be categorized as adhesive type. compounds on the frictional surface of steel
such as FeS, FeSO4, COH, COOH could be the
main mechanism in reducing friction (Xu et al.,
2007).
This study demonstrates that for each fuel,

higher wear occurs for higher speeds and the
(a) In B0 (c) In B50 increasing trends exactly follow the way of
increasing trends of friction. This may be
attributed to generation of higher heat between
the contact surfaces as a consequence of higher
speed which results higher wear. It is found that
pure biodiesel or higher biodiesel blends cause
less wear than that in lower biodiesel blends or
(b) In B20 (d) In B100 diesel. This indicates that the reduced wear
seems to happen due to presence of biodiesel.
Friction and wear in higher biodiesel blends are
Figure 7: Scanning electron micrographs (50X) more likely to be reduced by having a
of worn surfaces of used balls at speed 1500 comparatively more stable film to separate the
rpm. adjacent surfaces and limit metal to metal
contact. This can be attributed to the presence of
4. DISCUSSION aliphatic fatty acid of general formula
CnH2n+1COOH, such as stearic acid in POME
During run-in period for each speed, the wear which can enhance lubricating property by
rate and friction coefficient are initially quite controlling friction and wear between contact
high but as the contact surfaces become surfaces through developing of lubrication films
smoother and more prominent asperities are lost (Masjuki and Maleque, 1996). The protective
or flattened, the wear rate falls. With increasing films can reduce thermal energy in sliding
biodiesel concentration, the run in period for contact and thereby improve lubricity (Choi et
each speed is also reduced. This suggests that al., 1997).
biodiesel can successfully reduce the scuffling
period and thereby reduce the friction According to Brajendra et al. (2008) the ester
coefficient. It is found that the run-in period ends of the fatty acid chain are adsorbed to
decreases slightly with decreasing the speed. metal surfaces, thus permitting monolayer film
But the effect of each speed on steady state formation with the hydrocarbon end of fatty
friction coefficient for higher concentration acids oriented away from the metal surface. The
(>20%) of biodiesel are almost similar. fatty acid chain thus offers a sliding surface that
However, the average friction coefficient in prevents the direct metal-to-metal contact.
diesel is found to be higher than that in However, the stability of these films also
biodiesel. It is found that under a rotate speed of depends on operating conditions such as load,
1500 rpm, the average friction coefficients of temperature, speed as well as fluid viscosity and
diesel and biodiesel are 0.0857 and 0.0889 composition (Maleque et al., 2000). In an
respectively. This suggests that the films another explanation, trace components found in
provided by biodiesel at the contact surface are biodiesel fuels including free fatty acids,
more effective than that of diesel to reduce the monoglycerides, diglycerides are reported to
friction. According to Wain et al., (2005) improve the lubricity of biodiesel (Hu et al.,
biodiesel containing more oxygen can reduce 2005). Oxygenated moieties, degree of
friction as compared to diesel fuel. unsaturation, chain length of molecules in
biodiesel also play an important role in
Though the variation of friction coefficients for increasing lubricity. Geller and Goodrum,
different fuels is in small scale, it is found that (2004) have observed that lubricity is increased
the increasing trends for friction coefficient with with increasing of the degree of unsaturation.
speed are almost linear for B0 and B100 fuels They found that methyl linolenate (C18:3) in the
while for blends are not. This is more likely C18 series demonstrated the best performance
attributed to the non-uniform availability of the as a lubricity enhancing component and methyl
concentration of heteroatom (e.g. oxygen, stearate (C18:0) was the least effective. In an
sulfur) in different blends which plays important another explanation, unhindered electrons in the
role in creating the lubricating film. It was form of free electron pairs or double-bond
218
electrons toward the end of a chain of C atoms oxygenated moieties, degree of unsaturated
are likely to be effective in enhancing lubricity molecules, free fatty acid components etc. in
(Barbour et al., 2000). biodiesel.
Lubricity can also be influenced by oxidation (2) Lubricity in terms of wear and friction
process. This is because the oxidation process decreases with the increase of rotating speed.
reconverts esters into different fatty acids However, at the concentration of biodiesel
including formic acid, acetic acid, propionic above 20%, the steady state friction coefficients
acid, caproic acid etc. (Tsuchiya et al., 2006) are almost similar in each speed.
which seems to cause improved lubricity
(Haseeb et al., 2010c). Apparently, in short term (3) Deformation of the worn surfaces decreases
test, oxidation process can provide better with increasing the concentration of biodiesel in
lubricity but in case of long term application, it blends. The sample in pure biodiesel is
causes degradation of fuel and is therefore subjected to abrasive wear while in diesel it is
results in reduced lubricity, enhanced corrosion subjected by adhesive wear.
and degradation of materials (Tsuchiya et al.,
2006). So, it is assumed that effect of oxidation
ACKNOWLEDGEMENT
on lubricity is crucial. This possible mechanism
will be explored in future studies. The authors would like to acknowledge the
financial support provided by the Ministry of
Surface morphology of the worn scar shows that Science, Technology & Innovation (MOSTI)
layers of debris from the surfaces have been under the science fund by Grant No: 03-02-03-
extruded sideways, while flacks of debris are SF3073 and by the Institute of Research
extruded out from the contact interface in the Management and Consultancy, University of
sliding direction of the rotating ball. It is seen Malaya (UM) under the IPPP Fund Project No. :
that the edge of the surface at B0 has been PS093/2008B.
plastically deformed and elongated with more
cracks and wear debris. This deformation has
been gradually decreased in the following REFERENCES
samples in B20, B50. It seems that the cavities
Anastopoulos, G., Lois, E., Serdari, A., Zanikos,
are much bigger than 20μm and the particles
removed from the cavities are also more likely F., Stornas, S. and Kalligeros, S. 2001.
bigger than 20μm in size. According to the Lubrication Properties of Low-Sulfur Diesel
Fuels in the Presence of Specific Types of
definition given by Robinowicz (1994), removal
Fatty Acid Derivatives, Energy & Fuels, 15:
of wear debris bigger than 20μm results
106-112.
adhesive wear. This indicates that the wear for
Barbour, R.H., Rickerard, D.J. and Elliot, N.G.
sample in B0, B20, B50 can be categorized as
adhesive wear. But for the sample in B100, no 2000. Understanding Diesel Lubricity, SAE
such type of severe damages is seen. No cracks Tech. Paper No. 2000-01-1918.
Celik, I. and Aydin, O. 2011. Effects of B100
are found like other samples and also the edges
biodiesel on injector and pump piston,
are not so elongated and irregular. Some thin,
Tribology Transactions, 54: 421-431.
regular and radial scratches are seen on the
Choi, U.S., Ahn, B.G., Kwon, O.K. and Chun,
surface which result abrasive wear. Removal of
small particles (<20μm) with a few surface Y.J. 1997. Tribological behavior of some
features results abrasive wear (Sperring and antiwear additives in vegetable oils. PII:
S0301-679X(97)00039-X.
Nowell, 2005). This study shows that the
Demirbas, A. 2009. Biodiesel from waste
severity of the wear is reduced with increasing
cooking oil via base-catalytic and
the concentration of biodiesel in blends and
supercritical methanol transesterification,
better surface is found for pure biodiesel.
Energy Conversion and Management, 50:.
923–927.
Dimitrakis, W.J. 2003. The Importance of
5. CONCLUSIONS Lubricity. Hydrocarbon Eng., 8:37-39.
Fazal, M. A., Haseeb, A. S. M. A. and Masjuki,
The following conclusions can be drawn from H. H. 2011. Biodiesel feasibility study: An
this study: evaluation of material compatibility,
performance, emission and engine durability,
(1) Both wear and friction decrease with Renewable and Sustainable Energy Reviews,
increasing the concentration of biodiesel. This is 15:1329-1324.
more likely attributed to the presence of
219
Fraer, R., Dinh, H., Proc Kenneth, McCormick, methyl esters, Energy Conversion and
R.L., Chandler, K. and Buchholz, B. 2005. Management, 52: 108–116.
Operating Experience and Teardown Robinowicz, E. 1994. The least wear. Wear,
Analysis for Engines Operated on Biodiesel 100: 533–541.
Blends (B20), SAE Technical Paper No. Sharma, B. K., Doll, K.M. and Erhan, S. Z.
2005-01-3641. 2008. Ester hydroxy derivatives of methyl
Gan, S. and Kiat Ng H. 2010. Effects of oleate: Tribological, oxidation and low
antioxidant additives on pollutant formation temperature properties, Bioresource
from the combustion of palm oil methyl Technology 99: 7333–7340.
ester blends with diesel in a non-pressurised Shiroyama, H. 2007. The harmonization of
burner, Energy Conversion and automobile environmental standards
Management, 51: 1536–1546. between Japan, the United States and
Geller, D.P. and Goodrum, J.W. 2004. Effects Europe: the 'depoliticizing strategy by
of speciﬁc fatty acid methyl esters on diesel industry and the dynamics between firms
fuel lubricity, Fuel, 83:2351–2356. and governments in a transnational context.
Haseeb, A.S.M.A., Masjuki, H.H., Ann, L.J. The Pacific Review, 20: 351 – 370.
and Fazal, M.A. 2010a. Corrosion Sperring, T.P. and Nowell, T.J. 2005.
characteristics of copper and leaded bronze SYCLOPS—a qualitative debris
in palm biodiesel. Fuel Processing classiﬁcation system developed for RAF
Technology, 91: 329-334. early failure detection centres. Tribology
Haseeb, A.S.M.A., Masjuki, H.H., Siang, C.T. International, 38: 898–903.
and Fazal, M.A. 2010b. Compatibility of Tsuchiya, T., Shiotani, H., Goto, S., Sugiyama,
elastomers in palm biodiesel, Renewable G. and Maeda, A. 2006. Japanese Standards
Energy, 35: 2356-2361. for Diesel Fuel Containing 5% FAME
Haseeb, A.S.M.A. Sia, S.Y., Fazal, M.A. and blended diesel fuels and its Impact on
Masjuki, H.H. 2010c. Effect of temperature Corrosion, SAE Technical Paper No. 2006-
on tribological properties of palm biodiesel, 01-3303.
Energy, 35:1460–1464. Tung, S.C. and McMillan, M.L. 2004.
Has-imoglu, C., Ciniviz, M., Ozsert, I., Icingur, Automotive tribology overview of current
Y., Parlak, A. and Salman, M. S. 2008. advances and challenges for the future,
Performance characteristics of a low heat Tribology International, 37: 517-536.
rejection diesel engine operating with Wain, K.S., Perez, J.M., Chapman, E. and
biodiesel, Renewable Energy, 33: 1709– Boehman, A.L. 2005. Alternative and low
1715. sulfer fuel options: boundary lubrication
Hu, J., Du, Z., Li, C. and Min, E. 2005. Study performance and potential problems.
on the lubrication properties of biodiesel as Tribology International, 38: 313-319.
fuel lubricity enhancers. Fuel, 84: 1601-6. Xu, Y., Wang, Q., Hu, X. and Chen, J. 2007.
Karonis, D., Anostopoulos, G., Lois, E., Preliminary study on the tribological
Stournas, S., Zannikos, F. and Serdari, A. performance of straw based bio-fuel.
1999. Assessment of the Lubricity of Greek Procedings of the STLE/ASME Internation
Road Diesel and the Effect of the Addition Joint Tribology Conference: IJC2007, USA,
of Specific Types of Biodiesel. SAE pp. 1022-1024.
Technical Paper No. 1999-01-1471.
Maleque, M.A., Masjuki, H.H. and Haseeb,
A.S.M.A. 2000. Effect of Mechanical fators
on tribological properties of palm oil methyl
ester blended lubricant, wear, 239: 117-125.
Masjuki, H.H. and Maleque, M.A. 1997.
Investigation of the anti-wear characteristics
of palm oil methyl ester using a four ball-
ball tribotester test, Wear, 206: 179-186.
Masjuki, H.H. and Maleque, M.A. 1996. The
effect of palm oil diesel fuel contaminated
lubricant on sliding wear of cast irons
against mild steel, Wear, 198: 293-299.
Ozsezen, A. N. and Canakci, M. 2011.
Determination of performance and
combustion characteristics of a diesel engine
fueled with canola and waste palm oil
220

Arasu Paper PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Arasu Paper PDF

Uploaded by

Copyright:

Available Formats

Regional Tribology Conference

Bayview Hotel, Langkawi Island, Malaysia, 22-24 November 2011

Paper Reference ID: RTC 011

SURFACE TEMPERATURE DISTRIBUTION IN A COMPOSITE BRAKE ROTOR

A.A. Adebisi1, M. A. Maleque1 and Q. H Shah2

ABSTRACT brake rotor play an important role by influencing the

2.0 FINITE ELEMENT MODEL as shown in Figure 1. The FE model is developed

(a) Finite element model (b) Proton Wira brake assembly

Figure 1: Brake rotor assembly

Table 1: Description of the brake rotor assembly components

Rotor Solid 164 8787 7638

Pads Solid 164 133 640

master surface due to its stiffness, while the friction

where h is convection heat transfer coefficient, T  is

4.0 SIMULATION AND EXPERIMENTATION

The vehicle speed is 60km/h during the static running

Figure 4: Surface temperature distribution profile for 0.5MPa

Figure 5: Mid radial temperature distribution plot for 0.5MPa

Figure 6: Surface temperature distribution profile for 1.0MPa

Figure 7: Mid radial temperature distribution plot for 1.0MPa

Paper Reference ID: RTC 011

SURFACE TEMPERATURE DISTRIBUTION IN A COMPOSITE BRAKE ROTOR

A.A. Adebisi1, M. A. Maleque1 and Q. H Shah2

ABSTRACT brake rotor play an important role by influencing the

2.0 FINITE ELEMENT MODEL as shown in Figure 1. The FE model is developed

(a) Finite element model (b) Proton Wira brake assembly

Figure 1: Brake rotor assembly

Table 1: Description of the brake rotor assembly components

Rotor Solid 164 8787 7638

Pads Solid 164 133 640

master surface due to its stiffness, while the friction

where h is convection heat transfer coefficient, T  is

4.0 SIMULATION AND EXPERIMENTATION

The vehicle speed is 60km/h during the static running

Figure 4: Surface temperature distribution profile for 0.5MPa

Figure 5: Mid radial temperature distribution plot for 0.5MPa

Figure 6: Surface temperature distribution profile for 1.0MPa

Figure 7: Mid radial temperature distribution plot for 1.0MPa

Paper Reference ID: RTC 012

WEAR PROPERTIES OF Fe-C-Al CAST IRON AND CONVENTIONAL CAST

Department of Manufacturing and Materials Engineering

ABSTRACT field of machine parts and components such as

2.1 Casting Process

The moulding materials for casting are:

An induction furnace was used for melting the

Figure 2 Wear rate of Fe-C-Al and conventional

3.2 Wear Mechanism of Damage Surface

The wear worn or damage surface of both cast

However, worn surface of Fe-C-Al CI

oxidation occurs in that place and every moment

Figure 6 Friction coefficient of both

Ghaderi, A. R. Ahmadabadi, M. N. and

Hatate, M. Shiota, T. Takahashi, N. and

Keller, J. Fridrici, V. Kapsa, P. Vidaller, S. and

Krause, D. E., 1969, "Gray Iron-A Unique

Liu, X. Takamori, S. Osawa, Y. and Yin, F.

Prasad, B. K. 2007, Sliding wear response of

Riahi, A. R. and Alpas, A. T. 2003, Wear map

Terheci, M. Manory, R. R. and Hensler, J. H.

Xing, J. Wang, Q. and Chang, J. 2007,

Paper Reference ID: RTC 020

TRIBOLOGICAL STUDY OF Al-BASED NATURAL FIBRE REINFORCED

M. A. Malequea*, A. R. Zamania, R. J. Talibb